changes

improving devx :)

.github/ISSUE_TEMPLATE/bug-report.yml

@@ -8,16 +8,15 @@ body:
       value: |
         Thank you for taking the time to file a bug report.
 
-        Use this to report BUGS in LangChain. For usage questions, feature requests and general design questions, please use the [LangChain Forum](https://forum.langchain.com/).
+        For usage questions, feature requests and general design questions, please use the [LangChain Forum](https://forum.langchain.com/).
 
-        Relevant links to check before filing a bug report to see if your issue has already been reported, fixed or
-        if there's another way to solve your problem:
+        Check these before submitting to see if your issue has already been reported, fixed or if there's another way to solve your problem:
 
-        * [LangChain Forum](https://forum.langchain.com/),
-        * [LangChain documentation with the integrated search](https://docs.langchain.com/oss/python/langchain/overview),
-        * [API Reference](https://reference.langchain.com/python/),
+        * [Documentation](https://docs.langchain.com/oss/python/langchain/overview),
+        * [API Reference Documentation](https://reference.langchain.com/python/),
         * [LangChain ChatBot](https://chat.langchain.com/)
         * [GitHub search](https://github.com/langchain-ai/langchain),
+        * [LangChain Forum](https://forum.langchain.com/),
   - type: checkboxes
     id: checks
     attributes:
@@ -36,16 +35,48 @@ body:
           required: true
         - label: This is not related to the langchain-community package.
           required: true
-        - label: I read what a minimal reproducible example is (https://stackoverflow.com/help/minimal-reproducible-example).
-          required: true
         - label: I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.
           required: true
+  - type: checkboxes
+    id: package
+    attributes:
+      label: Package (Required)
+      description: |
+        Which `langchain` package(s) is this bug related to? Select at least one.
+
+        Note that if the package you are reporting for is not listed here, it is not in this repository (e.g. `langchain-google-genai` is in [`langchain-ai/langchain-google`](https://github.com/langchain-ai/langchain-google/)).
+
+        Please report issues for other packages to their respective repositories.
+      options:
+        - label: langchain
+        - label: langchain-openai
+        - label: langchain-anthropic
+        - label: langchain-classic
+        - label: langchain-core
+        - label: langchain-cli
+        - label: langchain-model-profiles
+        - label: langchain-tests
+        - label: langchain-text-splitters
+        - label: langchain-chroma
+        - label: langchain-deepseek
+        - label: langchain-exa
+        - label: langchain-fireworks
+        - label: langchain-groq
+        - label: langchain-huggingface
+        - label: langchain-mistralai
+        - label: langchain-nomic
+        - label: langchain-ollama
+        - label: langchain-perplexity
+        - label: langchain-prompty
+        - label: langchain-qdrant
+        - label: langchain-xai
+        - label: Other / not sure / general
   - type: textarea
     id: reproduction
     validations:
       required: true
     attributes:
-      label: Example Code
+      label: Example Code (Python)
       description: |
         Please add a self-contained, [minimal, reproducible, example](https://stackoverflow.com/help/minimal-reproducible-example) with your use case.
 
@@ -53,15 +84,12 @@ body:
 
         **Important!**
 
-        * Avoid screenshots when possible, as they are hard to read and (more importantly) don't allow others to copy-and-paste your code.
-        * Reduce your code to the minimum required to reproduce the issue if possible. This makes it much easier for others to help you.
-        * Use code tags (e.g., ```python ... ```) to correctly [format your code](https://help.github.com/en/github/writing-on-github/creating-and-highlighting-code-blocks#syntax-highlighting).
-        * INCLUDE the language label (e.g. `python`) after the first three backticks to enable syntax highlighting. (e.g., ```python rather than ```).
+        * Avoid screenshots, as they are hard to read and (more importantly) don't allow others to copy-and-paste your code.
+        * Reduce your code to the minimum required to reproduce the issue if possible.
 
+        (This will be automatically formatted into code, so no need for backticks.)
+      render: python
       placeholder: |
-        The following code:
-
-        ```python
         from langchain_core.runnables import RunnableLambda
 
         def bad_code(inputs) -> int:
@@ -69,17 +97,14 @@ body:
 
           chain = RunnableLambda(bad_code)
           chain.invoke('Hello!')
-        ```
   - type: textarea
-    id: error
-    validations:
-      required: false
     attributes:
       label: Error Message and Stack Trace (if applicable)
       description: |
-        If you are reporting an error, please include the full error message and stack trace.
-      placeholder: |
-        Exception + full stack trace
+        If you are reporting an error, please copy and paste the full error message and
+        stack trace.
+        (This will be automatically formatted into code, so no need for backticks.)
+      render: shell
   - type: textarea
     id: description
     attributes:
@@ -99,9 +124,7 @@ body:
     attributes:
       label: System Info
       description: |
-        Please share your system info with us. Do NOT skip this step and please don't trim
-        the output. Most users don't include enough information here and it makes it harder
-        for us to help you.
+        Please share your system info with us.
 
         Run the following command in your terminal and paste the output here:
 
@@ -113,8 +136,6 @@ body:
         from langchain_core import sys_info
         sys_info.print_sys_info()
         ```
-
-        alternatively, put the entire output of `pip freeze` here.
       placeholder: |
         python -m langchain_core.sys_info
     validations:

.github/ISSUE_TEMPLATE/config.yml

@@ -1,9 +1,18 @@
 blank_issues_enabled: false
 version: 2.1
 contact_links:
-  - name: 📚 Documentation
-    url: https://github.com/langchain-ai/docs/issues/new?template=langchain.yml
+  - name: 📚 Documentation issue
+    url: https://github.com/langchain-ai/docs/issues/new?template=01-langchain.yml
     about: Report an issue related to the LangChain documentation
   - name: 💬 LangChain Forum
     url:  https://forum.langchain.com/
     about: General community discussions and support
+  - name: 📚 LangChain Documentation
+    url: https://docs.langchain.com/oss/python/langchain/overview
+    about: View the official LangChain documentation
+  - name: 📚 API Reference Documentation
+    url: https://reference.langchain.com/python/
+    about: View the official LangChain API reference documentation
+  - name: 💬 LangChain Forum
+    url: https://forum.langchain.com/
+    about: Ask questions and get help from the community

.github/ISSUE_TEMPLATE/feature-request.yml

@@ -13,11 +13,11 @@ body:
         Relevant links to check before filing a feature request to see if your request has already been made or
         if there's another way to achieve what you want:
 
-        * [LangChain Forum](https://forum.langchain.com/),
-        * [LangChain documentation with the integrated search](https://docs.langchain.com/oss/python/langchain/overview),
-        * [API Reference](https://reference.langchain.com/python/),
+        * [Documentation](https://docs.langchain.com/oss/python/langchain/overview),
+        * [API Reference Documentation](https://reference.langchain.com/python/),
         * [LangChain ChatBot](https://chat.langchain.com/)
         * [GitHub search](https://github.com/langchain-ai/langchain),
+        * [LangChain Forum](https://forum.langchain.com/),
   - type: checkboxes
     id: checks
     attributes:
@@ -34,6 +34,40 @@ body:
           required: true
         - label: This is not related to the langchain-community package.
           required: true
+  - type: checkboxes
+    id: package
+    attributes:
+      label: Package (Required)
+      description: |
+        Which `langchain` package(s) is this request related to? Select at least one.
+
+        Note that if the package you are requesting for is not listed here, it is not in this repository (e.g. `langchain-google-genai` is in `langchain-ai/langchain`).
+
+        Please submit feature requests for other packages to their respective repositories.
+      options:
+        - label: langchain
+        - label: langchain-openai
+        - label: langchain-anthropic
+        - label: langchain-classic
+        - label: langchain-core
+        - label: langchain-cli
+        - label: langchain-model-profiles
+        - label: langchain-tests
+        - label: langchain-text-splitters
+        - label: langchain-chroma
+        - label: langchain-deepseek
+        - label: langchain-exa
+        - label: langchain-fireworks
+        - label: langchain-groq
+        - label: langchain-huggingface
+        - label: langchain-mistralai
+        - label: langchain-nomic
+        - label: langchain-ollama
+        - label: langchain-perplexity
+        - label: langchain-prompty
+        - label: langchain-qdrant
+        - label: langchain-xai
+        - label: Other / not sure / general
   - type: textarea
     id: feature-description
     validations:

.github/ISSUE_TEMPLATE/privileged.yml

@@ -18,3 +18,33 @@ body:
     attributes:
       label: Issue Content
       description: Add the content of the issue here.
+  - type: checkboxes
+    id: package
+    attributes:
+      label: Package (Required)
+      description: |
+        Please select package(s) that this issue is related to.
+      options:
+        - label: langchain
+        - label: langchain-openai
+        - label: langchain-anthropic
+        - label: langchain-classic
+        - label: langchain-core
+        - label: langchain-cli
+        - label: langchain-model-profiles
+        - label: langchain-tests
+        - label: langchain-text-splitters
+        - label: langchain-chroma
+        - label: langchain-deepseek
+        - label: langchain-exa
+        - label: langchain-fireworks
+        - label: langchain-groq
+        - label: langchain-huggingface
+        - label: langchain-mistralai
+        - label: langchain-nomic
+        - label: langchain-ollama
+        - label: langchain-perplexity
+        - label: langchain-prompty
+        - label: langchain-qdrant
+        - label: langchain-xai
+        - label: Other / not sure / general

.github/ISSUE_TEMPLATE/task.yml

@@ -25,13 +25,13 @@ body:
       label: Task Description
       description: |
         Provide a clear and detailed description of the task.
-        
+
         What needs to be done? Be specific about the scope and requirements.
       placeholder: |
         This task involves...
-        
+
         The goal is to...
-        
+
         Specific requirements:
         - ...
         - ...
@@ -43,7 +43,7 @@ body:
       label: Acceptance Criteria
       description: |
         Define the criteria that must be met for this task to be considered complete.
-        
+
         What are the specific deliverables or outcomes expected?
       placeholder: |
         This task will be complete when:
@@ -58,15 +58,15 @@ body:
       label: Context and Background
       description: |
         Provide any relevant context, background information, or links to related issues/PRs.
-        
+
         Why is this task needed? What problem does it solve?
       placeholder: |
         Background:
         - ...
-        
+
         Related issues/PRs:
         - #...
-        
+
         Additional context:
         - ...
     validations:
@@ -77,15 +77,45 @@ body:
       label: Dependencies
       description: |
         List any dependencies or blockers for this task.
-        
+
         Are there other tasks, issues, or external factors that need to be completed first?
       placeholder: |
         This task depends on:
         - [ ] Issue #...
         - [ ] PR #...
         - [ ] External dependency: ...
-        
+
         Blocked by:
         - ...
     validations:
       required: false
+  - type: checkboxes
+    id: package
+    attributes:
+      label: Package (Required)
+      description: |
+        Please select package(s) that this task is related to.
+      options:
+        - label: langchain
+        - label: langchain-openai
+        - label: langchain-anthropic
+        - label: langchain-classic
+        - label: langchain-core
+        - label: langchain-cli
+        - label: langchain-model-profiles
+        - label: langchain-tests
+        - label: langchain-text-splitters
+        - label: langchain-chroma
+        - label: langchain-deepseek
+        - label: langchain-exa
+        - label: langchain-fireworks
+        - label: langchain-groq
+        - label: langchain-huggingface
+        - label: langchain-mistralai
+        - label: langchain-nomic
+        - label: langchain-ollama
+        - label: langchain-perplexity
+        - label: langchain-prompty
+        - label: langchain-qdrant
+        - label: langchain-xai
+        - label: Other / not sure / general

.github/PULL_REQUEST_TEMPLATE.md

@@ -1,28 +1,30 @@
 (Replace this entire block of text)
 
-Thank you for contributing to LangChain! Follow these steps to mark your pull request as ready for review. **If any of these steps are not completed, your PR will not be considered for review.**
+Read the full contributing guidelines: https://docs.langchain.com/oss/python/contributing/overview
+
+Thank you for contributing to LangChain! Follow these steps to have your pull request considered as ready for review.
+
+1. PR title: Should follow the format: TYPE(SCOPE): DESCRIPTION
 
-- [ ] **PR title**: Follows the format: {TYPE}({SCOPE}): {DESCRIPTION}
   - Examples:
+    - fix(anthropic): resolve flag parsing error
     - feat(core): add multi-tenant support
-    - fix(cli): resolve flag parsing error
-    - docs(openai): update API usage examples
-  - Allowed `{TYPE}` values:
-    - feat, fix, docs, style, refactor, perf, test, build, ci, chore, revert, release
-  - Allowed `{SCOPE}` values (optional):
-    - core, cli, langchain, standard-tests, text-splitters, docs, anthropic, chroma, deepseek, exa, fireworks, groq, huggingface, mistralai, nomic, ollama, openai, perplexity, prompty, qdrant, xai, infra
-  - Once you've written the title, please delete this checklist item; do not include it in the PR.
+    - test(openai): update API usage tests
+  - Allowed TYPE and SCOPE values: https://github.com/langchain-ai/langchain/blob/master/.github/workflows/pr_lint.yml#L15-L33
 
-- [ ] **PR message**: ***Delete this entire checklist*** and replace with
-  - **Description:** a description of the change. Include a [closing keyword](https://docs.github.com/en/issues/tracking-your-work-with-issues/using-issues/linking-a-pull-request-to-an-issue#linking-a-pull-request-to-an-issue-using-a-keyword) if applicable to a relevant issue.
-  - **Issue:** the issue # it fixes, if applicable (e.g. Fixes #123)
-  - **Dependencies:** any dependencies required for this change
+2. PR description:
 
-- [ ] **Lint and test**: Run `make format`, `make lint` and `make test` from the root of the package(s) you've modified. **We will not consider a PR unless these three are passing in CI.** See [contribution guidelines](https://docs.langchain.com/oss/python/contributing) for more.
+  - Write 1-2 sentences summarizing the change.
+  - If this PR addresses a specific issue, please include "Fixes #ISSUE_NUMBER" in the description to automatically close the issue when the PR is merged.
+  - If there are any breaking changes, please clearly describe them.
+  - If this PR depends on another PR being merged first, please include "Depends on #PR_NUMBER" inthe description.
+
+3. Run `make format`, `make lint` and `make test` from the root of the package(s) you've modified.
+
+  - We will not consider a PR unless these three are passing in CI.
 
 Additional guidelines:
 
-- Most PRs should not touch more than one package.
-- Please do not add dependencies to `pyproject.toml` files (even optional ones) unless they are **required** for unit tests. Likewise, please do not update the `uv.lock` files unless you are adding a required dependency.
-- Changes should be backwards compatible.
-- Make sure optional dependencies are imported within a function.
+  - We ask that if you use generative AI for your contribution, you include a disclaimer.
+  - PRs should not touch more than one package unless absolutely necessary.
+  - Do not update the `uv.lock` files unless or add dependencies to `pyproject.toml` files (even optional ones) unless you have explicit permission to do so by a maintainer.

.github/actions/poetry_setup/action.yml

@@ -1,93 +0,0 @@
-# An action for setting up poetry install with caching.
-# Using a custom action since the default action does not
-# take poetry install groups into account.
-# Action code from:
-# https://github.com/actions/setup-python/issues/505#issuecomment-1273013236
-name: poetry-install-with-caching
-description: Poetry install with support for caching of dependency groups.
-
-inputs:
-  python-version:
-    description: Python version, supporting MAJOR.MINOR only
-    required: true
-
-  poetry-version:
-    description: Poetry version
-    required: true
-
-  cache-key:
-    description: Cache key to use for manual handling of caching
-    required: true
-
-  working-directory:
-    description: Directory whose poetry.lock file should be cached
-    required: true
-
-runs:
-  using: composite
-  steps:
-    - uses: actions/setup-python@v5
-      name: Setup python ${{ inputs.python-version }}
-      id: setup-python
-      with:
-        python-version: ${{ inputs.python-version }}
-
-    - uses: actions/cache@v4
-      id: cache-bin-poetry
-      name: Cache Poetry binary - Python ${{ inputs.python-version }}
-      env:
-        SEGMENT_DOWNLOAD_TIMEOUT_MIN: "1"
-      with:
-        path: |
-          /opt/pipx/venvs/poetry
-        # This step caches the poetry installation, so make sure it's keyed on the poetry version as well.
-        key: bin-poetry-${{ runner.os }}-${{ runner.arch }}-py-${{ inputs.python-version }}-${{ inputs.poetry-version }}
-
-    - name: Refresh shell hashtable and fixup softlinks
-      if: steps.cache-bin-poetry.outputs.cache-hit == 'true'
-      shell: bash
-      env:
-        POETRY_VERSION: ${{ inputs.poetry-version }}
-        PYTHON_VERSION: ${{ inputs.python-version }}
-      run: |
-        set -eux
-
-        # Refresh the shell hashtable, to ensure correct `which` output.
-        hash -r
-
-        # `actions/cache@v3` doesn't always seem able to correctly unpack softlinks.
-        # Delete and recreate the softlinks pipx expects to have.
-        rm /opt/pipx/venvs/poetry/bin/python
-        cd /opt/pipx/venvs/poetry/bin
-        ln -s "$(which "python$PYTHON_VERSION")" python
-        chmod +x python
-        cd /opt/pipx_bin/
-        ln -s /opt/pipx/venvs/poetry/bin/poetry poetry
-        chmod +x poetry
-
-        # Ensure everything got set up correctly.
-        /opt/pipx/venvs/poetry/bin/python --version
-        /opt/pipx_bin/poetry --version
-
-    - name: Install poetry
-      if: steps.cache-bin-poetry.outputs.cache-hit != 'true'
-      shell: bash
-      env:
-        POETRY_VERSION: ${{ inputs.poetry-version }}
-        PYTHON_VERSION: ${{ inputs.python-version }}
-      # Install poetry using the python version installed by setup-python step.
-      run: pipx install "poetry==$POETRY_VERSION" --python '${{ steps.setup-python.outputs.python-path }}' --verbose
-
-    - name: Restore pip and poetry cached dependencies
-      uses: actions/cache@v4
-      env:
-        SEGMENT_DOWNLOAD_TIMEOUT_MIN: "4"
-        WORKDIR: ${{ inputs.working-directory == '' && '.' || inputs.working-directory }}
-      with:
-        path: |
-          ~/.cache/pip
-          ~/.cache/pypoetry/virtualenvs
-          ~/.cache/pypoetry/cache
-          ~/.cache/pypoetry/artifacts
-          ${{ env.WORKDIR }}/.venv
-        key: py-deps-${{ runner.os }}-${{ runner.arch }}-py-${{ inputs.python-version }}-poetry-${{ inputs.poetry-version }}-${{ inputs.cache-key }}-${{ hashFiles(format('{0}/**/poetry.lock', env.WORKDIR)) }}

.github/pr-file-labeler.yml

@@ -7,13 +7,12 @@ core:
       - any-glob-to-any-file:
           - "libs/core/**/*"
 
-langchain:
+langchain-classic:
   - changed-files:
       - any-glob-to-any-file:
           - "libs/langchain/**/*"
-          - "libs/langchain_v1/**/*"
 
-v1:
+langchain:
   - changed-files:
       - any-glob-to-any-file:
           - "libs/langchain_v1/**/*"
@@ -28,6 +27,11 @@ standard-tests:
       - any-glob-to-any-file:
           - "libs/standard-tests/**/*"
 
+model-profiles:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/model-profiles/**/*"
+
 text-splitters:
   - changed-files:
       - any-glob-to-any-file:
@@ -39,6 +43,81 @@ integration:
       - any-glob-to-any-file:
           - "libs/partners/**/*"
 
+anthropic:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/anthropic/**/*"
+
+chroma:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/chroma/**/*"
+
+deepseek:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/deepseek/**/*"
+
+exa:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/exa/**/*"
+
+fireworks:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/fireworks/**/*"
+
+groq:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/groq/**/*"
+
+huggingface:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/huggingface/**/*"
+
+mistralai:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/mistralai/**/*"
+
+nomic:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/nomic/**/*"
+
+ollama:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/ollama/**/*"
+
+openai:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/openai/**/*"
+
+perplexity:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/perplexity/**/*"
+
+prompty:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/prompty/**/*"
+
+qdrant:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/qdrant/**/*"
+
+xai:
+  - changed-files:
+      - any-glob-to-any-file:
+          - "libs/partners/xai/**/*"
+
 # Infrastructure and DevOps
 infra:
   - changed-files:

.github/pr-title-labeler.yml

@@ -1,41 +0,0 @@
-# PR title labeler config
-#
-# Labels PRs based on conventional commit patterns in titles
-#
-# Format: type(scope): description or type!: description (breaking)
-
-add-missing-labels: true
-clear-prexisting: false
-include-commits: false
-include-title: true
-label-for-breaking-changes: breaking
-
-label-mapping:
-  documentation: ["docs"]
-  feature: ["feat"]
-  fix: ["fix"]
-  infra: ["build", "ci", "chore"]
-  integration:
-    [
-      "anthropic",
-      "chroma",
-      "deepseek",
-      "exa",
-      "fireworks",
-      "groq",
-      "huggingface",
-      "mistralai",
-      "nomic",
-      "ollama",
-      "openai",
-      "perplexity",
-      "prompty",
-      "qdrant",
-      "xai",
-    ]
-  linting: ["style"]
-  performance: ["perf"]
-  refactor: ["refactor"]
-  release: ["release"]
-  revert: ["revert"]
-  tests: ["test"]

.github/scripts/check_diff.py

@@ -25,19 +25,18 @@ import tomllib
 from get_min_versions import get_min_version_from_toml
 from packaging.requirements import Requirement
 
-LANGCHAIN_DIRS = [
-    "libs/core",
-    "libs/text-splitters",
-    "libs/langchain",
-    "libs/langchain_v1",
-    "libs/model-profiles",
-]
+# Define explicit dependency relationships for main LangChain packages.
+# Key = package directory, Value = set of directories that should trigger tests for this package
+PACKAGE_DEPENDENCIES: Dict[str, Set[str]] = {
+    "libs/core": set(),  # core has no upstream dependencies
+    "libs/text-splitters": {"libs/core"},
+    "libs/langchain": {"libs/core", "libs/text-splitters"},
+    "libs/langchain_v1": {"libs/core"},
+    "libs/model-profiles": set(),  # model-profiles is independent
+}
 
-# When set to True, we are ignoring core dependents
-# in order to be able to get CI to pass for each individual
-# package that depends on core
-# e.g. if you touch core, we don't then add textsplitters/etc to CI
-IGNORE_CORE_DEPENDENTS = False
+# All main LangChain directories (order doesn't matter with explicit deps)
+LANGCHAIN_DIRS = list(PACKAGE_DEPENDENCIES.keys())
 
 # ignored partners are removed from dependents
 # but still run if directly edited
@@ -61,7 +60,7 @@ def all_package_dirs() -> Set[str]:
 
 
 def dependents_graph() -> dict:
-    """Construct a mapping of package -> dependents
+    """Construct a mapping of package -> dependents.
 
     Done such that we can run tests on all dependents of a package when a change is made.
     """
@@ -126,6 +125,25 @@ def add_dependents(dirs_to_eval: Set[str], dependents: dict) -> List[str]:
     return list(updated)
 
 
+def get_affected_packages(changed_dir: str) -> Set[str]:
+    """Get all packages that should be tested when a directory changes.
+
+    Args:
+        changed_dir: The directory that was changed (e.g., "libs/core")
+
+    Returns:
+        Set of package directories that depend on the changed directory
+    """
+    affected = {changed_dir}
+
+    # Check each package to see if it depends on the changed directory
+    for pkg_dir, dependencies in PACKAGE_DEPENDENCIES.items():
+        if changed_dir in dependencies:
+            affected.add(pkg_dir)
+
+    return affected
+
+
 def _get_configs_for_single_dir(job: str, dir_: str) -> List[Dict[str, str]]:
     if job == "test-pydantic":
         return _get_pydantic_test_configs(dir_)
@@ -254,27 +272,23 @@ if __name__ == "__main__":
             # infrastructure don't inadvertently break package testing, even if the change
             # appears unrelated (e.g., documentation build workflows). This is intentionally
             # conservative to catch unexpected side effects from workflow modifications.
-            #
-            # Example: A PR modifying .github/workflows/api_doc_build.yml will trigger
-            # lint/test jobs for libs/core, libs/text-splitters, libs/langchain, and
-            # libs/langchain_v1, even though the workflow may only affect documentation.
             dirs_to_run["extended-test"].update(LANGCHAIN_DIRS)
 
         if file.startswith("libs/core"):
             dirs_to_run["codspeed"].add("libs/core")
-        if any(file.startswith(dir_) for dir_ in LANGCHAIN_DIRS):
-            # add that dir and all dirs after in LANGCHAIN_DIRS
-            # for extended testing
 
-            found = False
-            for dir_ in LANGCHAIN_DIRS:
-                if dir_ == "libs/core" and IGNORE_CORE_DEPENDENTS:
-                    dirs_to_run["extended-test"].add(dir_)
-                    continue
-                if file.startswith(dir_):
-                    found = True
-                if found:
-                    dirs_to_run["extended-test"].add(dir_)
+        # Check if file is in one of the main LangChain directories
+        matched_langchain_dir = None
+        for dir_ in LANGCHAIN_DIRS:
+            if file.startswith(dir_):
+                matched_langchain_dir = dir_
+                break
+
+        if matched_langchain_dir:
+            # Add the changed directory and all packages that depend on it
+            affected_packages = get_affected_packages(matched_langchain_dir)
+            dirs_to_run["extended-test"].update(affected_packages)
+
         elif file.startswith("libs/standard-tests"):
             # TODO: update to include all packages that rely on standard-tests (all partner packages)
             # Note: won't run on external repo partners

.github/scripts/get_min_versions.py

@@ -98,7 +98,7 @@ def _check_python_version_from_requirement(
         return True
     else:
         marker_str = str(requirement.marker)
-        if "python_version" or "python_full_version" in marker_str:
+        if "python_version" in marker_str or "python_full_version" in marker_str:
             python_version_str = "".join(
                 char
                 for char in marker_str

.github/workflows/_compile_integration_test.yml

@@ -35,7 +35,7 @@ jobs:
     timeout-minutes: 20
     name: "Python ${{ inputs.python-version }}"
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ inputs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"

.github/workflows/_lint.yml

@@ -38,7 +38,7 @@ jobs:
     timeout-minutes: 20
     steps:
       - name: "📋 Checkout Code"
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ inputs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"

.github/workflows/_release.yml

@@ -54,7 +54,7 @@ jobs:
       version: ${{ steps.check-version.outputs.version }}
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: Set up Python + uv
         uses: "./.github/actions/uv_setup"
@@ -105,7 +105,7 @@ jobs:
     outputs:
       release-body: ${{ steps.generate-release-body.outputs.release-body }}
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           repository: langchain-ai/langchain
           path: langchain
@@ -206,7 +206,7 @@ jobs:
       id-token: write
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - uses: actions/download-artifact@v6
         with:
@@ -237,7 +237,7 @@ jobs:
       contents: read
     timeout-minutes: 20
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       # We explicitly *don't* set up caching here. This ensures our tests are
       # maximally sensitive to catching breakage.
@@ -396,7 +396,7 @@ jobs:
       contents: read
     strategy:
       matrix:
-        partner: [openai]
+        partner: [anthropic]
       fail-fast: false # Continue testing other partners if one fails
     env:
       ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
@@ -412,7 +412,7 @@ jobs:
       AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT_NAME: ${{ secrets.AZURE_OPENAI_EMBEDDINGS_DEPLOYMENT_NAME }}
       LANGCHAIN_TESTS_USER_AGENT: ${{ secrets.LANGCHAIN_TESTS_USER_AGENT }}
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       # We implement this conditional as Github Actions does not have good support
       # for conditionally needing steps. https://github.com/actions/runner/issues/491
@@ -492,7 +492,7 @@ jobs:
         working-directory: ${{ inputs.working-directory }}
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: Set up Python + uv
         uses: "./.github/actions/uv_setup"
@@ -532,7 +532,7 @@ jobs:
         working-directory: ${{ inputs.working-directory }}
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: Set up Python + uv
         uses: "./.github/actions/uv_setup"

.github/workflows/_test.yml

@@ -33,7 +33,7 @@ jobs:
     name: "Python ${{ inputs.python-version }}"
     steps:
       - name: "📋 Checkout Code"
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ inputs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"

.github/workflows/_test_pydantic.yml

@@ -36,7 +36,7 @@ jobs:
     name: "Pydantic ~=${{ inputs.pydantic-version }}"
     steps:
       - name: "📋 Checkout Code"
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ inputs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"

.github/workflows/auto-label-by-package.yml

@@ -0,0 +1,107 @@
+name: Auto Label Issues by Package
+
+on:
+  issues:
+    types: [opened, edited]
+
+jobs:
+  label-by-package:
+    permissions:
+      issues: write
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Sync package labels
+        uses: actions/github-script@v8
+        with:
+          script: |
+            const body = context.payload.issue.body || "";
+
+            // Extract text under "### Package"
+            const match = body.match(/### Package\s+([\s\S]*?)\n###/i);
+            if (!match) return;
+
+            const packageSection = match[1].trim();
+
+            // Mapping table for package names to labels
+            const mapping = {
+              "langchain": "langchain",
+              "langchain-openai": "openai",
+              "langchain-anthropic": "anthropic",
+              "langchain-classic": "langchain-classic",
+              "langchain-core": "core",
+              "langchain-cli": "cli",
+              "langchain-model-profiles": "model-profiles",
+              "langchain-tests": "standard-tests",
+              "langchain-text-splitters": "text-splitters",
+              "langchain-chroma": "chroma",
+              "langchain-deepseek": "deepseek",
+              "langchain-exa": "exa",
+              "langchain-fireworks": "fireworks",
+              "langchain-groq": "groq",
+              "langchain-huggingface": "huggingface",
+              "langchain-mistralai": "mistralai",
+              "langchain-nomic": "nomic",
+              "langchain-ollama": "ollama",
+              "langchain-perplexity": "perplexity",
+              "langchain-prompty": "prompty",
+              "langchain-qdrant": "qdrant",
+              "langchain-xai": "xai",
+            };
+
+            // All possible package labels we manage
+            const allPackageLabels = Object.values(mapping);
+            const selectedLabels = [];
+
+            // Check if this is checkbox format (multiple selection)
+            const checkboxMatches = packageSection.match(/- \[x\]\s+([^\n\r]+)/gi);
+            if (checkboxMatches) {
+              // Handle checkbox format
+              for (const match of checkboxMatches) {
+                const packageName = match.replace(/- \[x\]\s+/i, '').trim();
+                const label = mapping[packageName];
+                if (label && !selectedLabels.includes(label)) {
+                  selectedLabels.push(label);
+                }
+              }
+            } else {
+              // Handle dropdown format (single selection)
+              const label = mapping[packageSection];
+              if (label) {
+                selectedLabels.push(label);
+              }
+            }
+
+            // Get current issue labels
+            const issue = await github.rest.issues.get({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              issue_number: context.issue.number
+            });
+
+            const currentLabels = issue.data.labels.map(label => label.name);
+            const currentPackageLabels = currentLabels.filter(label => allPackageLabels.includes(label));
+
+            // Determine labels to add and remove
+            const labelsToAdd = selectedLabels.filter(label => !currentPackageLabels.includes(label));
+            const labelsToRemove = currentPackageLabels.filter(label => !selectedLabels.includes(label));
+
+            // Add new labels
+            if (labelsToAdd.length > 0) {
+              await github.rest.issues.addLabels({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: context.issue.number,
+                labels: labelsToAdd
+              });
+            }
+
+            // Remove old labels
+            for (const label of labelsToRemove) {
+              await github.rest.issues.removeLabel({
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                issue_number: context.issue.number,
+                name: label
+              });
+            }

.github/workflows/check_core_versions.yml

@@ -18,7 +18,7 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: "✅ Verify pyproject.toml & version.py Match"
         run: |

.github/workflows/check_diffs.yml

@@ -47,7 +47,7 @@ jobs:
     if: ${{ !contains(github.event.pull_request.labels.*.name, 'ci-ignore') }}
     steps:
       - name: "📋 Checkout Code"
-        uses: actions/checkout@v5
+        uses: actions/checkout@v6
       - name: "🐍 Setup Python 3.11"
         uses: actions/setup-python@v6
         with:
@@ -141,7 +141,7 @@ jobs:
       run:
         working-directory: ${{ matrix.job-configs.working-directory }}
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: "🐍 Set up Python ${{ matrix.job-configs.python-version }} + UV"
         uses: "./.github/actions/uv_setup"
@@ -182,7 +182,7 @@ jobs:
         job-configs: ${{ fromJson(needs.build.outputs.codspeed) }}
       fail-fast: false
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
 
       - name: "📦 Install UV Package Manager"
         uses: astral-sh/setup-uv@v7

.github/workflows/integration_tests.yml

@@ -71,14 +71,14 @@ jobs:
         working-directory: ${{ fromJSON(needs.compute-matrix.outputs.matrix).working-directory }}
 
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           path: langchain
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           repository: langchain-ai/langchain-google
           path: langchain-google
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           repository: langchain-ai/langchain-aws
           path: langchain-aws

.github/workflows/pr_lint.yml

@@ -26,11 +26,13 @@
 #   * revert     — reverts a previous commit
 #   * release    — prepare a new release
 #
-# Allowed Scopes (optional):
-#   core, cli, langchain, langchain_v1, langchain-classic, standard-tests,
-#   text-splitters, docs, anthropic, chroma, deepseek, exa, fireworks, groq,
-#   huggingface, mistralai, nomic, ollama, openai, perplexity, prompty, qdrant,
-#   xai, infra, deps
+# Allowed Scope(s) (optional):
+#   core, cli, langchain, langchain_v1, langchain-classic, model-profiles,
+#   standard-tests, text-splitters, docs, anthropic, chroma, deepseek, exa,
+#   fireworks, groq, huggingface, mistralai, nomic, ollama, openai,
+#   perplexity, prompty, qdrant, xai, infra, deps
+#
+# Multiple scopes can be used by separating them with a comma.
 #
 # Rules:
 #   1. The 'Type' must start with a lowercase letter.
@@ -79,7 +81,6 @@ jobs:
             core
             cli
             langchain
-            langchain_v1
             langchain-classic
             model-profiles
             standard-tests
@@ -101,6 +102,7 @@ jobs:
             qdrant
             xai
             infra
+            deps
           requireScope: false
           disallowScopes: |
             release

.github/workflows/v03_api_doc_build.yml

@@ -23,12 +23,12 @@ jobs:
     permissions:
       contents: read
     steps:
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           ref: v0.3
           path: langchain
 
-      - uses: actions/checkout@v5
+      - uses: actions/checkout@v6
         with:
           repository: langchain-ai/langchain-api-docs-html
           path: langchain-api-docs-html

.gitignore

@@ -163,3 +163,6 @@ node_modules
 
 prof
 virtualenv/
+scratch/
+
+.langgraph_api/

AGENTS.md

@@ -1,255 +1,58 @@
-# Global Development Guidelines for LangChain Projects
+# Global development guidelines for the LangChain monorepo
 
-## Core Development Principles
+This document provides context to understand the LangChain Python project and assist with development.
 
-### 1. Maintain Stable Public Interfaces ⚠️ CRITICAL
+## Project architecture and context
 
-**Always attempt to preserve function signatures, argument positions, and names for exported/public methods.**
+### Monorepo structure
 
-❌ **Bad - Breaking Change:**
+This is a Python monorepo with multiple independently versioned packages that use `uv`.
 
-```python
-def get_user(id, verbose=False):  # Changed from `user_id`
-    pass
+```txt
+langchain/
+├── libs/
+│   ├── core/             # `langchain-core` primitives and base abstractions
+│   ├── langchain/        # `langchain-classic` (legacy, no new features)
+│   ├── langchain_v1/     # Actively maintained `langchain` package
+│   ├── partners/         # Third-party integrations
+│   │   ├── openai/       # OpenAI models and embeddings
+│   │   ├── anthropic/    # Anthropic (Claude) integration
+│   │   ├── ollama/       # Local model support
+│   │   └── ... (other integrations maintained by the LangChain team)
+│   ├── text-splitters/   # Document chunking utilities
+│   ├── standard-tests/   # Shared test suite for integrations
+│   ├── model-profiles/   # Model configuration profiles
+│   └── cli/              # Command-line interface tools
+├── .github/              # CI/CD workflows and templates
+├── .vscode/              # VSCode IDE standard settings and recommended extensions
+└── README.md             # Information about LangChain
 ```
 
-✅ **Good - Stable Interface:**
+- **Core layer** (`langchain-core`): Base abstractions, interfaces, and protocols. Users should not need to know about this layer directly.
+- **Implementation layer** (`langchain`): Concrete implementations and high-level public utilities
+- **Integration layer** (`partners/`): Third-party service integrations. Note that this monorepo is not exhaustive of all LangChain integrations; some are maintained in separate repos, such as `langchain-ai/langchain-google` and `langchain-ai/langchain-aws`. Usually these repos are cloned at the same level as this monorepo, so if needed, you can refer to their code directly by navigating to `../langchain-google/` from this monorepo.
+- **Testing layer** (`standard-tests/`): Standardized integration tests for partner integrations
 
-```python
-def get_user(user_id: str, verbose: bool = False) -> User:
-    """Retrieve user by ID with optional verbose output."""
-    pass
-```
+### Development tools & commands**
 
-**Before making ANY changes to public APIs:**
+- `uv` – Fast Python package installer and resolver (replaces pip/poetry)
+- `make` – Task runner for common development commands. Feel free to look at the `Makefile` for available commands and usage patterns.
+- `ruff` – Fast Python linter and formatter
+- `mypy` – Static type checking
+- `pytest` – Testing framework
 
-- Check if the function/class is exported in `__init__.py`
-- Look for existing usage patterns in tests and examples
-- Use keyword-only arguments for new parameters: `*, new_param: str = "default"`
-- Mark experimental features clearly with docstring warnings (using MkDocs Material admonitions, like `!!! warning`)
+This monorepo uses `uv` for dependency management. Local development uses editable installs: `[tool.uv.sources]`
 
-🧠 *Ask yourself:* "Would this change break someone's code if they used it last week?"
-
-### 2. Code Quality Standards
-
-**All Python code MUST include type hints and return types.**
-
-❌ **Bad:**
-
-```python
-def p(u, d):
-    return [x for x in u if x not in d]
-```
-
-✅ **Good:**
-
-```python
-def filter_unknown_users(users: list[str], known_users: set[str]) -> list[str]:
-    """Filter out users that are not in the known users set.
-
-    Args:
-        users: List of user identifiers to filter.
-        known_users: Set of known/valid user identifiers.
-
-    Returns:
-        List of users that are not in the known_users set.
-    """
-    return [user for user in users if user not in known_users]
-```
-
-**Style Requirements:**
-
-- Use descriptive, **self-explanatory variable names**. Avoid overly short or cryptic identifiers.
-- Attempt to break up complex functions (>20 lines) into smaller, focused functions where it makes sense
-- Avoid unnecessary abstraction or premature optimization
-- Follow existing patterns in the codebase you're modifying
-
-### 3. Testing Requirements
-
-**Every new feature or bugfix MUST be covered by unit tests.**
-
-**Test Organization:**
-
-- Unit tests: `tests/unit_tests/` (no network calls allowed)
-- Integration tests: `tests/integration_tests/` (network calls permitted)
-- Use `pytest` as the testing framework
-
-**Test Quality Checklist:**
-
-- [ ] Tests fail when your new logic is broken
-- [ ] Happy path is covered
-- [ ] Edge cases and error conditions are tested
-- [ ] Use fixtures/mocks for external dependencies
-- [ ] Tests are deterministic (no flaky tests)
-
-Checklist questions:
-
-- [ ] Does the test suite fail if your new logic is broken?
-- [ ] Are all expected behaviors exercised (happy path, invalid input, etc)?
-- [ ] Do tests use fixtures or mocks where needed?
-
-```python
-def test_filter_unknown_users():
-    """Test filtering unknown users from a list."""
-    users = ["alice", "bob", "charlie"]
-    known_users = {"alice", "bob"}
-
-    result = filter_unknown_users(users, known_users)
-
-    assert result == ["charlie"]
-    assert len(result) == 1
-```
-
-### 4. Security and Risk Assessment
-
-**Security Checklist:**
-
-- No `eval()`, `exec()`, or `pickle` on user-controlled input
-- Proper exception handling (no bare `except:`) and use a `msg` variable for error messages
-- Remove unreachable/commented code before committing
-- Race conditions or resource leaks (file handles, sockets, threads).
-- Ensure proper resource cleanup (file handles, connections)
-
-❌ **Bad:**
-
-```python
-def load_config(path):
-    with open(path) as f:
-        return eval(f.read())  # ⚠️ Never eval config
-```
-
-✅ **Good:**
-
-```python
-import json
-
-def load_config(path: str) -> dict:
-    with open(path) as f:
-        return json.load(f)
-```
-
-### 5. Documentation Standards
-
-**Use Google-style docstrings with Args section for all public functions.**
-
-❌ **Insufficient Documentation:**
-
-```python
-def send_email(to, msg):
-    """Send an email to a recipient."""
-```
-
-✅ **Complete Documentation:**
-
-```python
-def send_email(to: str, msg: str, *, priority: str = "normal") -> bool:
-    """
-    Send an email to a recipient with specified priority.
-
-    Args:
-        to: The email address of the recipient.
-        msg: The message body to send.
-        priority: Email priority level (`'low'`, `'normal'`, `'high'`).
-
-    Returns:
-        `True` if email was sent successfully, `False` otherwise.
-
-    Raises:
-        `InvalidEmailError`: If the email address format is invalid.
-        `SMTPConnectionError`: If unable to connect to email server.
-    """
-```
-
-**Documentation Guidelines:**
-
-- Types go in function signatures, NOT in docstrings
-  - If a default is present, DO NOT repeat it in the docstring unless there is post-processing or it is set conditionally.
-- Focus on "why" rather than "what" in descriptions
-- Document all parameters, return values, and exceptions
-- Keep descriptions concise but clear
-- Ensure American English spelling (e.g., "behavior", not "behaviour")
-
-📌 *Tip:* Keep descriptions concise but clear. Only document return values if non-obvious.
-
-### 6. Architectural Improvements
-
-**When you encounter code that could be improved, suggest better designs:**
-
-❌ **Poor Design:**
-
-```python
-def process_data(data, db_conn, email_client, logger):
-    # Function doing too many things
-    validated = validate_data(data)
-    result = db_conn.save(validated)
-    email_client.send_notification(result)
-    logger.log(f"Processed {len(data)} items")
-    return result
-```
-
-✅ **Better Design:**
-
-```python
-@dataclass
-class ProcessingResult:
-    """Result of data processing operation."""
-    items_processed: int
-    success: bool
-    errors: List[str] = field(default_factory=list)
-
-class DataProcessor:
-    """Handles data validation, storage, and notification."""
-
-    def __init__(self, db_conn: Database, email_client: EmailClient):
-        self.db = db_conn
-        self.email = email_client
-
-    def process(self, data: List[dict]) -> ProcessingResult:
-        """Process and store data with notifications."""
-        validated = self._validate_data(data)
-        result = self.db.save(validated)
-        self._notify_completion(result)
-        return result
-```
-
-**Design Improvement Areas:**
-
-If there's a **cleaner**, **more scalable**, or **simpler** design, highlight it and suggest improvements that would:
-
-- Reduce code duplication through shared utilities
-- Make unit testing easier
-- Improve separation of concerns (single responsibility)
-- Make unit testing easier through dependency injection
-- Add clarity without adding complexity
-- Prefer dataclasses for structured data
-
-## Development Tools & Commands
-
-### Package Management
-
-```bash
-# Add package
-uv add package-name
-
-# Sync project dependencies
-uv sync
-uv lock
-```
-
-### Testing
+Each package in `libs/` has its own `pyproject.toml` and `uv.lock`.
 
 ```bash
 # Run unit tests (no network)
 make test
 
-# Don't run integration tests, as API keys must be set
-
 # Run specific test file
 uv run --group test pytest tests/unit_tests/test_specific.py
 ```
 
-### Code Quality
-
 ```bash
 # Lint code
 make lint
@@ -261,66 +64,118 @@ make format
 uv run --group lint mypy .
 ```
 
-### Dependency Management Patterns
+#### Key config files
 
-**Local Development Dependencies:**
+- pyproject.toml: Main workspace configuration with dependency groups
+- uv.lock: Locked dependencies for reproducible builds
+- Makefile: Development tasks
 
-```toml
-[tool.uv.sources]
-langchain-core = { path = "../core", editable = true }
-langchain-tests = { path = "../standard-tests", editable = true }
-```
+#### Commit standards
 
-**For tools, use the `@tool` decorator from `langchain_core.tools`:**
+Suggest PR titles that follow Conventional Commits format. Refer to .github/workflows/pr_lint for allowed types and scopes.
 
-```python
-from langchain_core.tools import tool
+#### Pull request guidelines
 
-@tool
-def search_database(query: str) -> str:
-    """Search the database for relevant information.
+- Always add a disclaimer to the PR description mentioning how AI agents are involved with the contribution.
+- Describe the "why" of the changes, why the proposed solution is the right one. Limit prose.
+- Highlight areas of the proposed changes that require careful review.
+
+## Core development principles
+
+### Maintain stable public interfaces
+
+CRITICAL: Always attempt to preserve function signatures, argument positions, and names for exported/public methods. Do not make breaking changes.
+
+**Before making ANY changes to public APIs:**
+
+- Check if the function/class is exported in `__init__.py`
+- Look for existing usage patterns in tests and examples
+- Use keyword-only arguments for new parameters: `*, new_param: str = "default"`
+- Mark experimental features clearly with docstring warnings (using MkDocs Material admonitions, like `!!! warning`)
+
+Ask: "Would this change break someone's code if they used it last week?"
+
+### Code quality standards
+
+All Python code MUST include type hints and return types.
+
+```python title="Example"
+def filter_unknown_users(users: list[str], known_users: set[str]) -> list[str]:
+    """Single line description of the function.
+
+    Any additional context about the function can go here.
 
     Args:
-        query: The search query string.
+        users: List of user identifiers to filter.
+        known_users: Set of known/valid user identifiers.
+
+    Returns:
+        List of users that are not in the known_users set.
     """
-    # Implementation here
-    return results
 ```
 
-## Commit Standards
+- Use descriptive, self-explanatory variable names.
+- Follow existing patterns in the codebase you're modifying
+- Attempt to break up complex functions (>20 lines) into smaller, focused functions where it makes sense
 
-**Use Conventional Commits format for PR titles:**
+### Testing requirements
 
-- `feat(core): add multi-tenant support`
-- `fix(cli): resolve flag parsing error`
-- `docs: update API usage examples`
-- `docs(openai): update API usage examples`
+Every new feature or bugfix MUST be covered by unit tests.
 
-## Framework-Specific Guidelines
+- Unit tests: `tests/unit_tests/` (no network calls allowed)
+- Integration tests: `tests/integration_tests/` (network calls permitted)
+- We use `pytest` as the testing framework; if in doubt, check other existing tests for examples.
+- The testing file structure should mirror the source code structure.
 
-- Follow the existing patterns in `langchain-core` for base abstractions
-- Use `langchain_core.callbacks` for execution tracking
-- Implement proper streaming support where applicable
-- Avoid deprecated components like legacy `LLMChain`
+**Checklist:**
 
-### Partner Integrations
+- [ ] Tests fail when your new logic is broken
+- [ ] Happy path is covered
+- [ ] Edge cases and error conditions are tested
+- [ ] Use fixtures/mocks for external dependencies
+- [ ] Tests are deterministic (no flaky tests)
+- [ ] Does the test suite fail if your new logic is broken?
 
-- Follow the established patterns in existing partner libraries
-- Implement standard interfaces (`BaseChatModel`, `BaseEmbeddings`, etc.)
-- Include comprehensive integration tests
-- Document API key requirements and authentication
+### Security and risk assessment
 
----
+- No `eval()`, `exec()`, or `pickle` on user-controlled input
+- Proper exception handling (no bare `except:`) and use a `msg` variable for error messages
+- Remove unreachable/commented code before committing
+- Race conditions or resource leaks (file handles, sockets, threads).
+- Ensure proper resource cleanup (file handles, connections)
 
-## Quick Reference Checklist
+### Documentation standards
 
-Before submitting code changes:
+Use Google-style docstrings with Args section for all public functions.
 
-- [ ] **Breaking Changes**: Verified no public API changes
-- [ ] **Type Hints**: All functions have complete type annotations
-- [ ] **Tests**: New functionality is fully tested
-- [ ] **Security**: No dangerous patterns (eval, silent failures, etc.)
-- [ ] **Documentation**: Google-style docstrings for public functions
-- [ ] **Code Quality**: `make lint` and `make format` pass
-- [ ] **Architecture**: Suggested improvements where applicable
-- [ ] **Commit Message**: Follows Conventional Commits format
+```python title="Example"
+def send_email(to: str, msg: str, *, priority: str = "normal") -> bool:
+    """Send an email to a recipient with specified priority.
+
+    Any additional context about the function can go here.
+
+    Args:
+        to: The email address of the recipient.
+        msg: The message body to send.
+        priority: Email priority level.
+
+    Returns:
+        `True` if email was sent successfully, `False` otherwise.
+
+    Raises:
+        InvalidEmailError: If the email address format is invalid.
+        SMTPConnectionError: If unable to connect to email server.
+    """
+```
+
+- Types go in function signatures, NOT in docstrings
+  - If a default is present, DO NOT repeat it in the docstring unless there is post-processing or it is set conditionally.
+- Focus on "why" rather than "what" in descriptions
+- Document all parameters, return values, and exceptions
+- Keep descriptions concise but clear
+- Ensure American English spelling (e.g., "behavior", not "behaviour")
+
+## Additional resources
+
+- **Documentation:** https://docs.langchain.com/oss/python/langchain/overview and source at https://github.com/langchain-ai/docs or `../docs/`. Prefer the local install and use file search tools for best results. If needed, use the docs MCP server as defined in `.mcp.json` for programmatic access.
+- **Contributing Guide:** [`.github/CONTRIBUTING.md`](https://docs.langchain.com/oss/python/contributing/overview)

CLAUDE.md

@@ -1,255 +1,58 @@
-# Global Development Guidelines for LangChain Projects
+# Global development guidelines for the LangChain monorepo
 
-## Core Development Principles
+This document provides context to understand the LangChain Python project and assist with development.
 
-### 1. Maintain Stable Public Interfaces ⚠️ CRITICAL
+## Project architecture and context
 
-**Always attempt to preserve function signatures, argument positions, and names for exported/public methods.**
+### Monorepo structure
 
-❌ **Bad - Breaking Change:**
+This is a Python monorepo with multiple independently versioned packages that use `uv`.
 
-```python
-def get_user(id, verbose=False):  # Changed from `user_id`
-    pass
+```txt
+langchain/
+├── libs/
+│   ├── core/             # `langchain-core` primitives and base abstractions
+│   ├── langchain/        # `langchain-classic` (legacy, no new features)
+│   ├── langchain_v1/     # Actively maintained `langchain` package
+│   ├── partners/         # Third-party integrations
+│   │   ├── openai/       # OpenAI models and embeddings
+│   │   ├── anthropic/    # Anthropic (Claude) integration
+│   │   ├── ollama/       # Local model support
+│   │   └── ... (other integrations maintained by the LangChain team)
+│   ├── text-splitters/   # Document chunking utilities
+│   ├── standard-tests/   # Shared test suite for integrations
+│   ├── model-profiles/   # Model configuration profiles
+│   └── cli/              # Command-line interface tools
+├── .github/              # CI/CD workflows and templates
+├── .vscode/              # VSCode IDE standard settings and recommended extensions
+└── README.md             # Information about LangChain
 ```
 
-✅ **Good - Stable Interface:**
+- **Core layer** (`langchain-core`): Base abstractions, interfaces, and protocols. Users should not need to know about this layer directly.
+- **Implementation layer** (`langchain`): Concrete implementations and high-level public utilities
+- **Integration layer** (`partners/`): Third-party service integrations. Note that this monorepo is not exhaustive of all LangChain integrations; some are maintained in separate repos, such as `langchain-ai/langchain-google` and `langchain-ai/langchain-aws`. Usually these repos are cloned at the same level as this monorepo, so if needed, you can refer to their code directly by navigating to `../langchain-google/` from this monorepo.
+- **Testing layer** (`standard-tests/`): Standardized integration tests for partner integrations
 
-```python
-def get_user(user_id: str, verbose: bool = False) -> User:
-    """Retrieve user by ID with optional verbose output."""
-    pass
-```
+### Development tools & commands**
 
-**Before making ANY changes to public APIs:**
+- `uv` – Fast Python package installer and resolver (replaces pip/poetry)
+- `make` – Task runner for common development commands. Feel free to look at the `Makefile` for available commands and usage patterns.
+- `ruff` – Fast Python linter and formatter
+- `mypy` – Static type checking
+- `pytest` – Testing framework
 
-- Check if the function/class is exported in `__init__.py`
-- Look for existing usage patterns in tests and examples
-- Use keyword-only arguments for new parameters: `*, new_param: str = "default"`
-- Mark experimental features clearly with docstring warnings (using MkDocs Material admonitions, like `!!! warning`)
+This monorepo uses `uv` for dependency management. Local development uses editable installs: `[tool.uv.sources]`
 
-🧠 *Ask yourself:* "Would this change break someone's code if they used it last week?"
-
-### 2. Code Quality Standards
-
-**All Python code MUST include type hints and return types.**
-
-❌ **Bad:**
-
-```python
-def p(u, d):
-    return [x for x in u if x not in d]
-```
-
-✅ **Good:**
-
-```python
-def filter_unknown_users(users: list[str], known_users: set[str]) -> list[str]:
-    """Filter out users that are not in the known users set.
-
-    Args:
-        users: List of user identifiers to filter.
-        known_users: Set of known/valid user identifiers.
-
-    Returns:
-        List of users that are not in the known_users set.
-    """
-    return [user for user in users if user not in known_users]
-```
-
-**Style Requirements:**
-
-- Use descriptive, **self-explanatory variable names**. Avoid overly short or cryptic identifiers.
-- Attempt to break up complex functions (>20 lines) into smaller, focused functions where it makes sense
-- Avoid unnecessary abstraction or premature optimization
-- Follow existing patterns in the codebase you're modifying
-
-### 3. Testing Requirements
-
-**Every new feature or bugfix MUST be covered by unit tests.**
-
-**Test Organization:**
-
-- Unit tests: `tests/unit_tests/` (no network calls allowed)
-- Integration tests: `tests/integration_tests/` (network calls permitted)
-- Use `pytest` as the testing framework
-
-**Test Quality Checklist:**
-
-- [ ] Tests fail when your new logic is broken
-- [ ] Happy path is covered
-- [ ] Edge cases and error conditions are tested
-- [ ] Use fixtures/mocks for external dependencies
-- [ ] Tests are deterministic (no flaky tests)
-
-Checklist questions:
-
-- [ ] Does the test suite fail if your new logic is broken?
-- [ ] Are all expected behaviors exercised (happy path, invalid input, etc)?
-- [ ] Do tests use fixtures or mocks where needed?
-
-```python
-def test_filter_unknown_users():
-    """Test filtering unknown users from a list."""
-    users = ["alice", "bob", "charlie"]
-    known_users = {"alice", "bob"}
-
-    result = filter_unknown_users(users, known_users)
-
-    assert result == ["charlie"]
-    assert len(result) == 1
-```
-
-### 4. Security and Risk Assessment
-
-**Security Checklist:**
-
-- No `eval()`, `exec()`, or `pickle` on user-controlled input
-- Proper exception handling (no bare `except:`) and use a `msg` variable for error messages
-- Remove unreachable/commented code before committing
-- Race conditions or resource leaks (file handles, sockets, threads).
-- Ensure proper resource cleanup (file handles, connections)
-
-❌ **Bad:**
-
-```python
-def load_config(path):
-    with open(path) as f:
-        return eval(f.read())  # ⚠️ Never eval config
-```
-
-✅ **Good:**
-
-```python
-import json
-
-def load_config(path: str) -> dict:
-    with open(path) as f:
-        return json.load(f)
-```
-
-### 5. Documentation Standards
-
-**Use Google-style docstrings with Args section for all public functions.**
-
-❌ **Insufficient Documentation:**
-
-```python
-def send_email(to, msg):
-    """Send an email to a recipient."""
-```
-
-✅ **Complete Documentation:**
-
-```python
-def send_email(to: str, msg: str, *, priority: str = "normal") -> bool:
-    """
-    Send an email to a recipient with specified priority.
-
-    Args:
-        to: The email address of the recipient.
-        msg: The message body to send.
-        priority: Email priority level (`'low'`, `'normal'`, `'high'`).
-
-    Returns:
-        `True` if email was sent successfully, `False` otherwise.
-
-    Raises:
-        `InvalidEmailError`: If the email address format is invalid.
-        `SMTPConnectionError`: If unable to connect to email server.
-    """
-```
-
-**Documentation Guidelines:**
-
-- Types go in function signatures, NOT in docstrings
-  - If a default is present, DO NOT repeat it in the docstring unless there is post-processing or it is set conditionally.
-- Focus on "why" rather than "what" in descriptions
-- Document all parameters, return values, and exceptions
-- Keep descriptions concise but clear
-- Ensure American English spelling (e.g., "behavior", not "behaviour")
-
-📌 *Tip:* Keep descriptions concise but clear. Only document return values if non-obvious.
-
-### 6. Architectural Improvements
-
-**When you encounter code that could be improved, suggest better designs:**
-
-❌ **Poor Design:**
-
-```python
-def process_data(data, db_conn, email_client, logger):
-    # Function doing too many things
-    validated = validate_data(data)
-    result = db_conn.save(validated)
-    email_client.send_notification(result)
-    logger.log(f"Processed {len(data)} items")
-    return result
-```
-
-✅ **Better Design:**
-
-```python
-@dataclass
-class ProcessingResult:
-    """Result of data processing operation."""
-    items_processed: int
-    success: bool
-    errors: List[str] = field(default_factory=list)
-
-class DataProcessor:
-    """Handles data validation, storage, and notification."""
-
-    def __init__(self, db_conn: Database, email_client: EmailClient):
-        self.db = db_conn
-        self.email = email_client
-
-    def process(self, data: List[dict]) -> ProcessingResult:
-        """Process and store data with notifications."""
-        validated = self._validate_data(data)
-        result = self.db.save(validated)
-        self._notify_completion(result)
-        return result
-```
-
-**Design Improvement Areas:**
-
-If there's a **cleaner**, **more scalable**, or **simpler** design, highlight it and suggest improvements that would:
-
-- Reduce code duplication through shared utilities
-- Make unit testing easier
-- Improve separation of concerns (single responsibility)
-- Make unit testing easier through dependency injection
-- Add clarity without adding complexity
-- Prefer dataclasses for structured data
-
-## Development Tools & Commands
-
-### Package Management
-
-```bash
-# Add package
-uv add package-name
-
-# Sync project dependencies
-uv sync
-uv lock
-```
-
-### Testing
+Each package in `libs/` has its own `pyproject.toml` and `uv.lock`.
 
 ```bash
 # Run unit tests (no network)
 make test
 
-# Don't run integration tests, as API keys must be set
-
 # Run specific test file
 uv run --group test pytest tests/unit_tests/test_specific.py
 ```
 
-### Code Quality
-
 ```bash
 # Lint code
 make lint
@@ -261,66 +64,118 @@ make format
 uv run --group lint mypy .
 ```
 
-### Dependency Management Patterns
+#### Key config files
 
-**Local Development Dependencies:**
+- pyproject.toml: Main workspace configuration with dependency groups
+- uv.lock: Locked dependencies for reproducible builds
+- Makefile: Development tasks
 
-```toml
-[tool.uv.sources]
-langchain-core = { path = "../core", editable = true }
-langchain-tests = { path = "../standard-tests", editable = true }
-```
+#### Commit standards
 
-**For tools, use the `@tool` decorator from `langchain_core.tools`:**
+Suggest PR titles that follow Conventional Commits format. Refer to .github/workflows/pr_lint for allowed types and scopes.
 
-```python
-from langchain_core.tools import tool
+#### Pull request guidelines
 
-@tool
-def search_database(query: str) -> str:
-    """Search the database for relevant information.
+- Always add a disclaimer to the PR description mentioning how AI agents are involved with the contribution.
+- Describe the "why" of the changes, why the proposed solution is the right one. Limit prose.
+- Highlight areas of the proposed changes that require careful review.
+
+## Core development principles
+
+### Maintain stable public interfaces
+
+CRITICAL: Always attempt to preserve function signatures, argument positions, and names for exported/public methods. Do not make breaking changes.
+
+**Before making ANY changes to public APIs:**
+
+- Check if the function/class is exported in `__init__.py`
+- Look for existing usage patterns in tests and examples
+- Use keyword-only arguments for new parameters: `*, new_param: str = "default"`
+- Mark experimental features clearly with docstring warnings (using MkDocs Material admonitions, like `!!! warning`)
+
+Ask: "Would this change break someone's code if they used it last week?"
+
+### Code quality standards
+
+All Python code MUST include type hints and return types.
+
+```python title="Example"
+def filter_unknown_users(users: list[str], known_users: set[str]) -> list[str]:
+    """Single line description of the function.
+
+    Any additional context about the function can go here.
 
     Args:
-        query: The search query string.
+        users: List of user identifiers to filter.
+        known_users: Set of known/valid user identifiers.
+
+    Returns:
+        List of users that are not in the known_users set.
     """
-    # Implementation here
-    return results
 ```
 
-## Commit Standards
+- Use descriptive, self-explanatory variable names.
+- Follow existing patterns in the codebase you're modifying
+- Attempt to break up complex functions (>20 lines) into smaller, focused functions where it makes sense
 
-**Use Conventional Commits format for PR titles:**
+### Testing requirements
 
-- `feat(core): add multi-tenant support`
-- `fix(cli): resolve flag parsing error`
-- `docs: update API usage examples`
-- `docs(openai): update API usage examples`
+Every new feature or bugfix MUST be covered by unit tests.
 
-## Framework-Specific Guidelines
+- Unit tests: `tests/unit_tests/` (no network calls allowed)
+- Integration tests: `tests/integration_tests/` (network calls permitted)
+- We use `pytest` as the testing framework; if in doubt, check other existing tests for examples.
+- The testing file structure should mirror the source code structure.
 
-- Follow the existing patterns in `langchain-core` for base abstractions
-- Use `langchain_core.callbacks` for execution tracking
-- Implement proper streaming support where applicable
-- Avoid deprecated components like legacy `LLMChain`
+**Checklist:**
 
-### Partner Integrations
+- [ ] Tests fail when your new logic is broken
+- [ ] Happy path is covered
+- [ ] Edge cases and error conditions are tested
+- [ ] Use fixtures/mocks for external dependencies
+- [ ] Tests are deterministic (no flaky tests)
+- [ ] Does the test suite fail if your new logic is broken?
 
-- Follow the established patterns in existing partner libraries
-- Implement standard interfaces (`BaseChatModel`, `BaseEmbeddings`, etc.)
-- Include comprehensive integration tests
-- Document API key requirements and authentication
+### Security and risk assessment
 
----
+- No `eval()`, `exec()`, or `pickle` on user-controlled input
+- Proper exception handling (no bare `except:`) and use a `msg` variable for error messages
+- Remove unreachable/commented code before committing
+- Race conditions or resource leaks (file handles, sockets, threads).
+- Ensure proper resource cleanup (file handles, connections)
 
-## Quick Reference Checklist
+### Documentation standards
 
-Before submitting code changes:
+Use Google-style docstrings with Args section for all public functions.
 
-- [ ] **Breaking Changes**: Verified no public API changes
-- [ ] **Type Hints**: All functions have complete type annotations
-- [ ] **Tests**: New functionality is fully tested
-- [ ] **Security**: No dangerous patterns (eval, silent failures, etc.)
-- [ ] **Documentation**: Google-style docstrings for public functions
-- [ ] **Code Quality**: `make lint` and `make format` pass
-- [ ] **Architecture**: Suggested improvements where applicable
-- [ ] **Commit Message**: Follows Conventional Commits format
+```python title="Example"
+def send_email(to: str, msg: str, *, priority: str = "normal") -> bool:
+    """Send an email to a recipient with specified priority.
+
+    Any additional context about the function can go here.
+
+    Args:
+        to: The email address of the recipient.
+        msg: The message body to send.
+        priority: Email priority level.
+
+    Returns:
+        `True` if email was sent successfully, `False` otherwise.
+
+    Raises:
+        InvalidEmailError: If the email address format is invalid.
+        SMTPConnectionError: If unable to connect to email server.
+    """
+```
+
+- Types go in function signatures, NOT in docstrings
+  - If a default is present, DO NOT repeat it in the docstring unless there is post-processing or it is set conditionally.
+- Focus on "why" rather than "what" in descriptions
+- Document all parameters, return values, and exceptions
+- Keep descriptions concise but clear
+- Ensure American English spelling (e.g., "behavior", not "behaviour")
+
+## Additional resources
+
+- **Documentation:** https://docs.langchain.com/oss/python/langchain/overview and source at https://github.com/langchain-ai/docs or `../docs/`. Prefer the local install and use file search tools for best results. If needed, use the docs MCP server as defined in `.mcp.json` for programmatic access.
+- **Contributing Guide:** [`.github/CONTRIBUTING.md`](https://docs.langchain.com/oss/python/contributing/overview)

MIGRATE.md

@@ -1,9 +0,0 @@
-# Migrating
-
-Please see the following guides for migrating LangChain code:
-
-* Migrate to [LangChain v1.0](https://docs.langchain.com/oss/python/migrate/langchain-v1)
-* Migrate to [LangChain v0.3](https://python.langchain.com/docs/versions/v0_3/)
-* Migrate to [LangChain v0.2](https://python.langchain.com/docs/versions/v0_2/)
-* Migrating from [LangChain 0.0.x Chains](https://python.langchain.com/docs/versions/migrating_chains/)
-* Upgrade to [LangGraph Memory](https://python.langchain.com/docs/versions/migrating_memory/)

README.md

@@ -1,40 +1,28 @@
-<p align="center">
-  <picture>
-    <source media="(prefers-color-scheme: light)" srcset=".github/images/logo-dark.svg">
-    <source media="(prefers-color-scheme: dark)" srcset=".github/images/logo-light.svg">
-    <img alt="LangChain Logo" src=".github/images/logo-dark.svg" width="80%">
-  </picture>
-</p>
+<div align="center">
+  <a href="https://www.langchain.com/">
+    <picture>
+      <source media="(prefers-color-scheme: light)" srcset=".github/images/logo-dark.svg">
+      <source media="(prefers-color-scheme: dark)" srcset=".github/images/logo-light.svg">
+      <img alt="LangChain Logo" src=".github/images/logo-dark.svg" width="80%">
+    </picture>
+  </a>
+</div>
 
-<p align="center">
-    The platform for reliable agents.
-</p>
+<div align="center">
+  <h3>The platform for reliable agents.</h3>
+</div>
 
-<p align="center">
-  <a href="https://opensource.org/licenses/MIT" target="_blank">
-      <img src="https://img.shields.io/pypi/l/langchain" alt="PyPI - License">
-  </a>
-  <a href="https://pypistats.org/packages/langchain" target="_blank">
-      <img src="https://img.shields.io/pepy/dt/langchain" alt="PyPI - Downloads">
-  </a>
-  <a href="https://pypi.org/project/langchain/#history" target="_blank">
-      <img src="https://img.shields.io/pypi/v/langchain?label=%20" alt="Version">
-  </a>
-  <a href="https://vscode.dev/redirect?url=vscode://ms-vscode-remote.remote-containers/cloneInVolume?url=https://github.com/langchain-ai/langchain" target="_blank">
-      <img src="https://img.shields.io/static/v1?label=Dev%20Containers&message=Open&color=blue&logo=visualstudiocode" alt="Open in Dev Containers">
-  </a>
-  <a href="https://codespaces.new/langchain-ai/langchain" target="_blank">
-      <img src="https://github.com/codespaces/badge.svg" alt="Open in Github Codespace" title="Open in Github Codespace" width="150" height="20">
-  </a>
-  <a href="https://codspeed.io/langchain-ai/langchain" target="_blank">
-      <img src="https://img.shields.io/endpoint?url=https://codspeed.io/badge.json" alt="CodSpeed Badge">
-  </a>
-  <a href="https://twitter.com/langchainai" target="_blank">
-      <img src="https://img.shields.io/twitter/url/https/twitter.com/langchainai.svg?style=social&label=Follow%20%40LangChainAI" alt="Twitter / X">
-  </a>
-</p>
+<div align="center">
+  <a href="https://opensource.org/licenses/MIT" target="_blank"><img src="https://img.shields.io/pypi/l/langchain" alt="PyPI - License"></a>
+  <a href="https://pypistats.org/packages/langchain" target="_blank"><img src="https://img.shields.io/pepy/dt/langchain" alt="PyPI - Downloads"></a>
+  <a href="https://pypi.org/project/langchain/#history" target="_blank"><img src="https://img.shields.io/pypi/v/langchain?label=%20" alt="Version"></a>
+  <a href="https://vscode.dev/redirect?url=vscode://ms-vscode-remote.remote-containers/cloneInVolume?url=https://github.com/langchain-ai/langchain" target="_blank"><img src="https://img.shields.io/static/v1?label=Dev%20Containers&message=Open&color=blue&logo=visualstudiocode" alt="Open in Dev Containers"></a>
+  <a href="https://codespaces.new/langchain-ai/langchain" target="_blank"><img src="https://github.com/codespaces/badge.svg" alt="Open in Github Codespace" title="Open in Github Codespace" width="150" height="20"></a>
+  <a href="https://codspeed.io/langchain-ai/langchain" target="_blank"><img src="https://img.shields.io/endpoint?url=https://codspeed.io/badge.json" alt="CodSpeed Badge"></a>
+  <a href="https://twitter.com/langchainai" target="_blank"><img src="https://img.shields.io/twitter/url/https/twitter.com/langchainai.svg?style=social&label=Follow%20%40LangChainAI" alt="Twitter / X"></a>
+</div>
 
-LangChain is a framework for building agents and LLM-powered applications. It helps you chain together interoperable components and third-party integrations to simplify AI application development —  all while future-proofing decisions as the underlying technology evolves.
+LangChain is a framework for building agents and LLM-powered applications. It helps you chain together interoperable components and third-party integrations to simplify AI application development – all while future-proofing decisions as the underlying technology evolves.
 
 ```bash
 pip install langchain
@@ -44,7 +32,10 @@ If you're looking for more advanced customization or agent orchestration, check
 
 ---
 
-**Documentation**: To learn more about LangChain, check out [the docs](https://docs.langchain.com/oss/python/langchain/overview).
+**Documentation**:
+
+- [docs.langchain.com](https://docs.langchain.com/oss/python/langchain/overview) – Comprehensive documentation, including conceptual overviews and guides
+- [reference.langchain.com/python](https://reference.langchain.com/python) – API reference docs for LangChain packages
 
 **Discussions**: Visit the [LangChain Forum](https://forum.langchain.com) to connect with the community and share all of your technical questions, ideas, and feedback.
 
@@ -57,8 +48,12 @@ LangChain helps developers build applications powered by LLMs through a standard
 
 Use LangChain for:
 
-- **Real-time data augmentation**. Easily connect LLMs to diverse data sources and external/internal systems, drawing from LangChain’s vast library of integrations with model providers, tools, vector stores, retrievers, and more.
-- **Model interoperability**. Swap models in and out as your engineering team experiments to find the best choice for your application’s needs. As the industry frontier evolves, adapt quickly — LangChain’s abstractions keep you moving without losing momentum.
+- **Real-time data augmentation**. Easily connect LLMs to diverse data sources and external/internal systems, drawing from LangChain's vast library of integrations with model providers, tools, vector stores, retrievers, and more.
+- **Model interoperability**. Swap models in and out as your engineering team experiments to find the best choice for your application's needs. As the industry frontier evolves, adapt quickly – LangChain's abstractions keep you moving without losing momentum.
+- **Rapid prototyping**. Quickly build and iterate on LLM applications with LangChain's modular, component-based architecture. Test different approaches and workflows without rebuilding from scratch, accelerating your development cycle.
+- **Production-ready features**. Deploy reliable applications with built-in support for monitoring, evaluation, and debugging through integrations like LangSmith. Scale with confidence using battle-tested patterns and best practices.
+- **Vibrant community and ecosystem**. Leverage a rich ecosystem of integrations, templates, and community-contributed components. Benefit from continuous improvements and stay up-to-date with the latest AI developments through an active open-source community.
+- **Flexible abstraction layers**. Work at the level of abstraction that suits your needs - from high-level chains for quick starts to low-level components for fine-grained control. LangChain grows with your application's complexity.
 
 ## LangChain ecosystem
 
@@ -66,13 +61,14 @@ While the LangChain framework can be used standalone, it also integrates seamles
 
 To improve your LLM application development, pair LangChain with:
 
-- [LangGraph](https://docs.langchain.com/oss/python/langgraph/overview) - Build agents that can reliably handle complex tasks with LangGraph, our low-level agent orchestration framework. LangGraph offers customizable architecture, long-term memory, and human-in-the-loop workflows — and is trusted in production by companies like LinkedIn, Uber, Klarna, and GitLab.
-- [LangSmith](https://www.langchain.com/langsmith) - Helpful for agent evals and observability. Debug poor-performing LLM app runs, evaluate agent trajectories, gain visibility in production, and improve performance over time.
-- [LangSmith Deployment](https://docs.langchain.com/langsmith/deployments) - Deploy and scale agents effortlessly with a purpose-built deployment platform for long-running, stateful workflows. Discover, reuse, configure, and share agents across teams — and iterate quickly with visual prototyping in [LangSmith Studio](https://docs.langchain.com/langsmith/studio).
+- [LangGraph](https://docs.langchain.com/oss/python/langgraph/overview) – Build agents that can reliably handle complex tasks with LangGraph, our low-level agent orchestration framework. LangGraph offers customizable architecture, long-term memory, and human-in-the-loop workflows – and is trusted in production by companies like LinkedIn, Uber, Klarna, and GitLab.
+- [Integrations](https://docs.langchain.com/oss/python/integrations/providers/overview) – List of LangChain integrations, including chat & embedding models, tools & toolkits, and more
+- [LangSmith](https://www.langchain.com/langsmith) – Helpful for agent evals and observability. Debug poor-performing LLM app runs, evaluate agent trajectories, gain visibility in production, and improve performance over time.
+- [LangSmith Deployment](https://docs.langchain.com/langsmith/deployments) – Deploy and scale agents effortlessly with a purpose-built deployment platform for long-running, stateful workflows. Discover, reuse, configure, and share agents across teams – and iterate quickly with visual prototyping in [LangSmith Studio](https://docs.langchain.com/langsmith/studio).
+- [Deep Agents](https://github.com/langchain-ai/deepagents) *(new!)* – Build agents that can plan, use subagents, and leverage file systems for complex tasks
 
 ## Additional resources
 
-- [API Reference](https://reference.langchain.com/python): Detailed reference on navigating base packages and integrations for LangChain.
-- [Integrations](https://docs.langchain.com/oss/python/integrations/providers/overview): List of LangChain integrations, including chat & embedding models, tools & toolkits, and more
-- [Contributing Guide](https://docs.langchain.com/oss/python/contributing/overview): Learn how to contribute to LangChain and find good first issues.
-- [Code of Conduct](https://github.com/langchain-ai/langchain/blob/master/.github/CODE_OF_CONDUCT.md): Our community guidelines and standards for participation.
+- [API Reference](https://reference.langchain.com/python) – Detailed reference on navigating base packages and integrations for LangChain.
+- [Contributing Guide](https://docs.langchain.com/oss/python/contributing/overview) – Learn how to contribute to LangChain projects and find good first issues.
+- [Code of Conduct](https://github.com/langchain-ai/langchain/blob/master/.github/CODE_OF_CONDUCT.md) – Our community guidelines and standards for participation.

SECURITY.md

@@ -55,10 +55,10 @@ All out of scope targets defined by huntr as well as:
 * **langchain-experimental**: This repository is for experimental code and is not
   eligible for bug bounties (see [package warning](https://pypi.org/project/langchain-experimental/)), bug reports to it will be marked as interesting or waste of
   time and published with no bounty attached.
-* **tools**: Tools in either langchain or langchain-community are not eligible for bug
+* **tools**: Tools in either `langchain` or `langchain-community` are not eligible for bug
   bounties. This includes the following directories
-  * libs/langchain/langchain/tools
-  * libs/community/langchain_community/tools
+  * `libs/langchain/langchain/tools`
+  * `libs/community/langchain_community/tools`
   * Please review the [Best Practices](#best-practices)
     for more details, but generally tools interact with the real world. Developers are
     expected to understand the security implications of their code and are responsible

libs/cli/langchain_cli/integration_template/docs/vectorstores.ipynb

@@ -295,7 +295,7 @@
    "source": [
     "## TODO: Any functionality specific to this vector store\n",
     "\n",
-    "E.g. creating a persisten database to save to your disk, etc."
+    "E.g. creating a persistent database to save to your disk, etc."
    ]
   },
   {

libs/cli/langchain_cli/utils/git.py

@@ -6,9 +6,8 @@ import hashlib
 import logging
 import re
 import shutil
-from collections.abc import Sequence
 from pathlib import Path
-from typing import Any, TypedDict
+from typing import TYPE_CHECKING, Any, TypedDict
 
 from git import Repo
 
@@ -18,6 +17,9 @@ from langchain_cli.constants import (
     DEFAULT_GIT_SUBDIRECTORY,
 )
 
+if TYPE_CHECKING:
+    from collections.abc import Sequence
+
 logger = logging.getLogger(__name__)
 
 

libs/cli/tests/unit_tests/migrate/cli_runner/case.py

@@ -1,9 +1,11 @@
 from __future__ import annotations
 
 from dataclasses import dataclass
+from typing import TYPE_CHECKING
 
-from .file import File
-from .folder import Folder
+if TYPE_CHECKING:
+    from .file import File
+    from .folder import Folder
 
 
 @dataclass

libs/cli/tests/unit_tests/migrate/cli_runner/folder.py

@@ -1,9 +1,12 @@
 from __future__ import annotations
 
-from pathlib import Path
+from typing import TYPE_CHECKING
 
 from .file import File
 
+if TYPE_CHECKING:
+    from pathlib import Path
+
 
 class Folder:
     def __init__(self, name: str, *files: Folder | File) -> None:

libs/core/README.md

@@ -34,7 +34,7 @@ The LangChain ecosystem is built on top of `langchain-core`. Some of the benefit
 
 ## 📖 Documentation
 
-For full documentation, see the [API reference](https://reference.langchain.com/python/langchain_core/).
+For full documentation, see the [API reference](https://reference.langchain.com/python/langchain_core/). For conceptual guides, tutorials, and examples on using LangChain, see the [LangChain Docs](https://docs.langchain.com/oss/python/langchain/overview).
 
 ## 📕 Releases & Versioning
 

libs/core/langchain_core/agents.py

@@ -52,31 +52,33 @@ class AgentAction(Serializable):
     """The input to pass in to the Tool."""
     log: str
     """Additional information to log about the action.
-    This log can be used in a few ways. First, it can be used to audit
-    what exactly the LLM predicted to lead to this (tool, tool_input).
-    Second, it can be used in future iterations to show the LLMs prior
-    thoughts. This is useful when (tool, tool_input) does not contain
-    full information about the LLM prediction (for example, any `thought`
-    before the tool/tool_input)."""
+
+    This log can be used in a few ways. First, it can be used to audit what exactly the
+    LLM predicted to lead to this `(tool, tool_input)`.
+
+    Second, it can be used in future iterations to show the LLMs prior thoughts. This is
+    useful when `(tool, tool_input)` does not contain full information about the LLM
+    prediction (for example, any `thought` before the tool/tool_input).
+    """
     type: Literal["AgentAction"] = "AgentAction"
 
     # Override init to support instantiation by position for backward compat.
     def __init__(self, tool: str, tool_input: str | dict, log: str, **kwargs: Any):
-        """Create an AgentAction.
+        """Create an `AgentAction`.
 
         Args:
             tool: The name of the tool to execute.
-            tool_input: The input to pass in to the Tool.
+            tool_input: The input to pass in to the `Tool`.
             log: Additional information to log about the action.
         """
         super().__init__(tool=tool, tool_input=tool_input, log=log, **kwargs)
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """AgentAction is serializable.
+        """`AgentAction` is serializable.
 
         Returns:
-            True
+            `True`
         """
         return True
 
@@ -98,19 +100,23 @@ class AgentAction(Serializable):
 class AgentActionMessageLog(AgentAction):
     """Representation of an action to be executed by an agent.
 
-    This is similar to AgentAction, but includes a message log consisting of
-    chat messages. This is useful when working with ChatModels, and is used
-    to reconstruct conversation history from the agent's perspective.
+    This is similar to `AgentAction`, but includes a message log consisting of
+    chat messages.
+
+    This is useful when working with `ChatModels`, and is used to reconstruct
+    conversation history from the agent's perspective.
     """
 
     message_log: Sequence[BaseMessage]
-    """Similar to log, this can be used to pass along extra
-    information about what exact messages were predicted by the LLM
-    before parsing out the (tool, tool_input). This is again useful
-    if (tool, tool_input) cannot be used to fully recreate the LLM
-    prediction, and you need that LLM prediction (for future agent iteration).
+    """Similar to log, this can be used to pass along extra information about what exact
+    messages were predicted by the LLM before parsing out the `(tool, tool_input)`.
+
+    This is again useful if `(tool, tool_input)` cannot be used to fully recreate the
+    LLM prediction, and you need that LLM prediction (for future agent iteration).
+
     Compared to `log`, this is useful when the underlying LLM is a
-    chat model (and therefore returns messages rather than a string)."""
+    chat model (and therefore returns messages rather than a string).
+    """
     # Ignoring type because we're overriding the type from AgentAction.
     # And this is the correct thing to do in this case.
     # The type literal is used for serialization purposes.
@@ -132,19 +138,22 @@ class AgentStep(Serializable):
 
 
 class AgentFinish(Serializable):
-    """Final return value of an ActionAgent.
+    """Final return value of an `ActionAgent`.
 
-    Agents return an AgentFinish when they have reached a stopping condition.
+    Agents return an `AgentFinish` when they have reached a stopping condition.
     """
 
     return_values: dict
     """Dictionary of return values."""
     log: str
     """Additional information to log about the return value.
+
     This is used to pass along the full LLM prediction, not just the parsed out
-    return value. For example, if the full LLM prediction was
-    `Final Answer: 2` you may want to just return `2` as a return value, but pass
-    along the full string as a `log` (for debugging or observability purposes).
+    return value.
+
+    For example, if the full LLM prediction was `Final Answer: 2` you may want to just
+    return `2` as a return value, but pass along the full string as a `log` (for
+    debugging or observability purposes).
     """
     type: Literal["AgentFinish"] = "AgentFinish"
 
@@ -154,7 +163,7 @@ class AgentFinish(Serializable):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -202,7 +211,7 @@ def _convert_agent_observation_to_messages(
         observation: Observation to convert to a message.
 
     Returns:
-        AIMessage that corresponds to the original tool invocation.
+        `AIMessage` that corresponds to the original tool invocation.
     """
     if isinstance(agent_action, AgentActionMessageLog):
         return [_create_function_message(agent_action, observation)]
@@ -225,7 +234,7 @@ def _create_function_message(
         observation: the result of the tool invocation.
 
     Returns:
-        FunctionMessage that corresponds to the original tool invocation.
+        `FunctionMessage` that corresponds to the original tool invocation.
     """
     if not isinstance(observation, str):
         try:

libs/core/langchain_core/callbacks/base.py

@@ -5,13 +5,12 @@ from __future__ import annotations
 import logging
 from typing import TYPE_CHECKING, Any
 
-from typing_extensions import Self
-
 if TYPE_CHECKING:
     from collections.abc import Sequence
     from uuid import UUID
 
     from tenacity import RetryCallState
+    from typing_extensions import Self
 
     from langchain_core.agents import AgentAction, AgentFinish
     from langchain_core.documents import Document

libs/core/langchain_core/callbacks/manager.py

@@ -39,7 +39,6 @@ from langchain_core.tracers.context import (
     tracing_v2_callback_var,
 )
 from langchain_core.tracers.langchain import LangChainTracer
-from langchain_core.tracers.schemas import Run
 from langchain_core.tracers.stdout import ConsoleCallbackHandler
 from langchain_core.utils.env import env_var_is_set
 
@@ -52,6 +51,7 @@ if TYPE_CHECKING:
     from langchain_core.documents import Document
     from langchain_core.outputs import ChatGenerationChunk, GenerationChunk, LLMResult
     from langchain_core.runnables.config import RunnableConfig
+    from langchain_core.tracers.schemas import Run
 
 logger = logging.getLogger(__name__)
 
@@ -229,7 +229,24 @@ def shielded(func: Func) -> Func:
 
     @functools.wraps(func)
     async def wrapped(*args: Any, **kwargs: Any) -> Any:
-        return await asyncio.shield(func(*args, **kwargs))
+        # Capture the current context to preserve context variables
+        ctx = copy_context()
+
+        # Create the coroutine
+        coro = func(*args, **kwargs)
+
+        # For Python 3.11+, create task with explicit context
+        # For older versions, fallback to original behavior
+        try:
+            # Create a task with the captured context to preserve context variables
+            task = asyncio.create_task(coro, context=ctx)  # type: ignore[call-arg, unused-ignore]
+            # `call-arg` used to not fail 3.9 or 3.10 tests
+            return await asyncio.shield(task)
+        except TypeError:
+            # Python < 3.11 fallback - create task normally then shield
+            # This won't preserve context perfectly but is better than nothing
+            task = asyncio.create_task(coro)
+            return await asyncio.shield(task)
 
     return cast("Func", wrapped)
 

libs/core/langchain_core/callbacks/usage.py

@@ -43,7 +43,7 @@ class UsageMetadataCallbackHandler(BaseCallbackHandler):
           'input_token_details': {'cache_read': 0, 'cache_creation': 0}}}
         ```
 
-    !!! version-added "Added in version 0.3.49"
+    !!! version-added "Added in `langchain-core` 0.3.49"
 
     """
 
@@ -134,7 +134,7 @@ def get_usage_metadata_callback(
         }
         ```
 
-    !!! version-added "Added in version 0.3.49"
+    !!! version-added "Added in `langchain-core` 0.3.49"
 
     """
     usage_metadata_callback_var: ContextVar[UsageMetadataCallbackHandler | None] = (

libs/core/langchain_core/documents/base.py

@@ -114,11 +114,11 @@ class Blob(BaseMedia):
     data: bytes | str | None = None
     """Raw data associated with the `Blob`."""
     mimetype: str | None = None
-    """MimeType not to be confused with a file extension."""
+    """MIME type, not to be confused with a file extension."""
     encoding: str = "utf-8"
     """Encoding to use if decoding the bytes into a string.
 
-    Use `utf-8` as default encoding, if decoding to string.
+    Uses `utf-8` as default encoding if decoding to string.
     """
     path: PathLike | None = None
     """Location where the original content was found."""
@@ -134,7 +134,7 @@ class Blob(BaseMedia):
 
         If a path is associated with the `Blob`, it will default to the path location.
 
-        Unless explicitly set via a metadata field called `"source"`, in which
+        Unless explicitly set via a metadata field called `'source'`, in which
         case that value will be used instead.
         """
         if self.metadata and "source" in self.metadata:
@@ -309,7 +309,7 @@ class Document(BaseMedia):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -322,10 +322,10 @@ class Document(BaseMedia):
         return ["langchain", "schema", "document"]
 
     def __str__(self) -> str:
-        """Override __str__ to restrict it to page_content and metadata.
+        """Override `__str__` to restrict it to page_content and metadata.
 
         Returns:
-            A string representation of the Document.
+            A string representation of the `Document`.
         """
         # The format matches pydantic format for __str__.
         #

libs/core/langchain_core/example_selectors/length_based.py

@@ -29,7 +29,7 @@ class LengthBasedExampleSelector(BaseExampleSelector, BaseModel):
     max_length: int = 2048
     """Max length for the prompt, beyond which examples are cut."""
 
-    example_text_lengths: list[int] = Field(default_factory=list)  # :meta private:
+    example_text_lengths: list[int] = Field(default_factory=list)
     """Length of each example."""
 
     def add_example(self, example: dict[str, str]) -> None:

libs/core/langchain_core/indexing/api.py

@@ -6,16 +6,9 @@ import hashlib
 import json
 import uuid
 import warnings
-from collections.abc import (
-    AsyncIterable,
-    AsyncIterator,
-    Callable,
-    Iterable,
-    Iterator,
-    Sequence,
-)
 from itertools import islice
 from typing import (
+    TYPE_CHECKING,
     Any,
     Literal,
     TypedDict,
@@ -29,6 +22,16 @@ from langchain_core.exceptions import LangChainException
 from langchain_core.indexing.base import DocumentIndex, RecordManager
 from langchain_core.vectorstores import VectorStore
 
+if TYPE_CHECKING:
+    from collections.abc import (
+        AsyncIterable,
+        AsyncIterator,
+        Callable,
+        Iterable,
+        Iterator,
+        Sequence,
+    )
+
 # Magic UUID to use as a namespace for hashing.
 # Used to try and generate a unique UUID for each document
 # from hashing the document content and metadata.
@@ -298,7 +301,8 @@ def index(
     For the time being, documents are indexed using their hashes, and users
     are not able to specify the uid of the document.
 
-    !!! warning "Behavior changed in 0.3.25"
+    !!! warning "Behavior changed in `langchain-core` 0.3.25"
+
         Added `scoped_full` cleanup mode.
 
     !!! warning
@@ -349,7 +353,7 @@ def index(
         key_encoder: Hashing algorithm to use for hashing the document content and
             metadata. Options include "blake2b", "sha256", and "sha512".
 
-            !!! version-added "Added in version 0.3.66"
+            !!! version-added "Added in `langchain-core` 0.3.66"
 
         key_encoder: Hashing algorithm to use for hashing the document.
             If not provided, a default encoder using SHA-1 will be used.
@@ -366,7 +370,7 @@ def index(
             method of the `VectorStore` or the upsert method of the DocumentIndex.
             For example, you can use this to specify a custom vector_field:
             upsert_kwargs={"vector_field": "embedding"}
-            !!! version-added "Added in version 0.3.10"
+            !!! version-added "Added in `langchain-core` 0.3.10"
 
     Returns:
         Indexing result which contains information about how many documents
@@ -636,7 +640,8 @@ async def aindex(
     For the time being, documents are indexed using their hashes, and users
     are not able to specify the uid of the document.
 
-    !!! warning "Behavior changed in 0.3.25"
+    !!! warning "Behavior changed in `langchain-core` 0.3.25"
+
         Added `scoped_full` cleanup mode.
 
     !!! warning
@@ -687,7 +692,7 @@ async def aindex(
         key_encoder: Hashing algorithm to use for hashing the document content and
             metadata. Options include "blake2b", "sha256", and "sha512".
 
-            !!! version-added "Added in version 0.3.66"
+            !!! version-added "Added in `langchain-core` 0.3.66"
 
         key_encoder: Hashing algorithm to use for hashing the document.
             If not provided, a default encoder using SHA-1 will be used.
@@ -704,7 +709,7 @@ async def aindex(
             method of the `VectorStore` or the upsert method of the DocumentIndex.
             For example, you can use this to specify a custom vector_field:
             upsert_kwargs={"vector_field": "embedding"}
-            !!! version-added "Added in version 0.3.10"
+            !!! version-added "Added in `langchain-core` 0.3.10"
 
     Returns:
         Indexing result which contains information about how many documents

libs/core/langchain_core/language_models/__init__.py

@@ -53,6 +53,10 @@ if TYPE_CHECKING:
         ParrotFakeChatModel,
     )
     from langchain_core.language_models.llms import LLM, BaseLLM
+    from langchain_core.language_models.model_profile import (
+        ModelProfile,
+        ModelProfileRegistry,
+    )
 
 __all__ = (
     "LLM",
@@ -68,6 +72,8 @@ __all__ = (
     "LanguageModelInput",
     "LanguageModelLike",
     "LanguageModelOutput",
+    "ModelProfile",
+    "ModelProfileRegistry",
     "ParrotFakeChatModel",
     "SimpleChatModel",
     "get_tokenizer",
@@ -90,6 +96,8 @@ _dynamic_imports = {
     "GenericFakeChatModel": "fake_chat_models",
     "ParrotFakeChatModel": "fake_chat_models",
     "LLM": "llms",
+    "ModelProfile": "model_profile",
+    "ModelProfileRegistry": "model_profile",
     "BaseLLM": "llms",
     "is_openai_data_block": "_utils",
 }

libs/core/langchain_core/language_models/_utils.py

@@ -139,7 +139,8 @@ def _normalize_messages(
             directly; this may change in the future
     - LangChain v0 standard content blocks for backward compatibility
 
-    !!! warning "Behavior changed in 1.0.0"
+    !!! warning "Behavior changed in `langchain-core` 1.0.0"
+
         In previous versions, this function returned messages in LangChain v0 format.
         Now, it returns messages in LangChain v1 format, which upgraded chat models now
         expect to receive when passing back in message history. For backward

libs/core/langchain_core/language_models/base.py

@@ -131,14 +131,19 @@ class BaseLanguageModel(
 
     Caching is not currently supported for streaming methods of models.
     """
+
     verbose: bool = Field(default_factory=_get_verbosity, exclude=True, repr=False)
     """Whether to print out response text."""
+
     callbacks: Callbacks = Field(default=None, exclude=True)
     """Callbacks to add to the run trace."""
+
     tags: list[str] | None = Field(default=None, exclude=True)
     """Tags to add to the run trace."""
+
     metadata: dict[str, Any] | None = Field(default=None, exclude=True)
     """Metadata to add to the run trace."""
+
     custom_get_token_ids: Callable[[str], list[int]] | None = Field(
         default=None, exclude=True
     )
@@ -195,15 +200,22 @@ class BaseLanguageModel(
             type (e.g., pure text completion models vs chat models).
 
         Args:
-            prompts: List of `PromptValue` objects. A `PromptValue` is an object that
-                can be converted to match the format of any language model (string for
-                pure text generation models and `BaseMessage` objects for chat models).
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            prompts: List of `PromptValue` objects.
+
+                A `PromptValue` is an object that can be converted to match the format
+                of any language model (string for pure text generation models and
+                `BaseMessage` objects for chat models).
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             An `LLMResult`, which contains a list of candidate `Generation` objects for
@@ -232,15 +244,22 @@ class BaseLanguageModel(
             type (e.g., pure text completion models vs chat models).
 
         Args:
-            prompts: List of `PromptValue` objects. A `PromptValue` is an object that
-                can be converted to match the format of any language model (string for
-                pure text generation models and `BaseMessage` objects for chat models).
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            prompts: List of `PromptValue` objects.
+
+                A `PromptValue` is an object that can be converted to match the format
+                of any language model (string for pure text generation models and
+                `BaseMessage` objects for chat models).
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             An `LLMResult`, which contains a list of candidate `Generation` objects for
@@ -280,6 +299,9 @@ class BaseLanguageModel(
 
         Useful for checking if an input fits in a model's context window.
 
+        This should be overridden by model-specific implementations to provide accurate
+        token counts via model-specific tokenizers.
+
         Args:
             text: The string input to tokenize.
 
@@ -298,9 +320,17 @@ class BaseLanguageModel(
 
         Useful for checking if an input fits in a model's context window.
 
+        This should be overridden by model-specific implementations to provide accurate
+        token counts via model-specific tokenizers.
+
         !!! note
-            The base implementation of `get_num_tokens_from_messages` ignores tool
-            schemas.
+
+            * The base implementation of `get_num_tokens_from_messages` ignores tool
+                schemas.
+            * The base implementation of `get_num_tokens_from_messages` adds additional
+                prefixes to messages in represent user roles, which will add to the
+                overall token count. Model-specific implementations may choose to
+                handle this differently.
 
         Args:
             messages: The message inputs to tokenize.

libs/core/langchain_core/language_models/chat_models.py

@@ -15,7 +15,6 @@ from typing import TYPE_CHECKING, Any, Literal, cast
 from pydantic import BaseModel, ConfigDict, Field
 from typing_extensions import override
 
-from langchain_core._api.beta_decorator import beta
 from langchain_core.caches import BaseCache
 from langchain_core.callbacks import (
     AsyncCallbackManager,
@@ -34,6 +33,7 @@ from langchain_core.language_models.base import (
     LangSmithParams,
     LanguageModelInput,
 )
+from langchain_core.language_models.model_profile import ModelProfile
 from langchain_core.load import dumpd, dumps
 from langchain_core.messages import (
     AIMessage,
@@ -76,8 +76,6 @@ from langchain_core.utils.utils import LC_ID_PREFIX, from_env
 if TYPE_CHECKING:
     import uuid
 
-    from langchain_model_profiles import ModelProfile  # type: ignore[import-untyped]
-
     from langchain_core.output_parsers.base import OutputParserLike
     from langchain_core.runnables import Runnable, RunnableConfig
     from langchain_core.tools import BaseTool
@@ -91,7 +89,10 @@ def _generate_response_from_error(error: BaseException) -> list[ChatGeneration]:
             try:
                 metadata["body"] = response.json()
             except Exception:
-                metadata["body"] = getattr(response, "text", None)
+                try:
+                    metadata["body"] = getattr(response, "text", None)
+                except Exception:
+                    metadata["body"] = None
         if hasattr(response, "headers"):
             try:
                 metadata["headers"] = dict(response.headers)
@@ -332,10 +333,25 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
     [`langchain-openai`](https://pypi.org/project/langchain-openai)) can also use this
     field to roll out new content formats in a backward-compatible way.
 
-    !!! version-added "Added in version 1.0"
+    !!! version-added "Added in `langchain-core` 1.0.0"
 
     """
 
+    profile: ModelProfile | None = Field(default=None, exclude=True)
+    """Profile detailing model capabilities.
+
+    !!! warning "Beta feature"
+        This is a beta feature. The format of model profiles is subject to change.
+
+    If not specified, automatically loaded from the provider package on initialization
+    if data is available.
+
+    Example profile data includes context window sizes, supported modalities, or support
+    for tool calling, structured output, and other features.
+
+    !!! version-added "Added in `langchain-core` 1.1.0"
+    """
+
     model_config = ConfigDict(
         arbitrary_types_allowed=True,
     )
@@ -845,16 +861,21 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
 
         Args:
             messages: List of list of messages.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
             tags: The tags to apply.
             metadata: The metadata to apply.
             run_name: The name of the run.
             run_id: The ID of the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             An `LLMResult`, which contains a list of candidate `Generations` for each
@@ -963,16 +984,21 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
 
         Args:
             messages: List of list of messages.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
             tags: The tags to apply.
             metadata: The metadata to apply.
             run_name: The name of the run.
             run_id: The ID of the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             An `LLMResult`, which contains a list of candidate `Generations` for each
@@ -1505,10 +1531,10 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
         Args:
             schema: The output schema. Can be passed in as:
 
-                - an OpenAI function/tool schema,
-                - a JSON Schema,
-                - a `TypedDict` class,
-                - or a Pydantic class.
+                - An OpenAI function/tool schema,
+                - A JSON Schema,
+                - A `TypedDict` class,
+                - Or a Pydantic class.
 
                 If `schema` is a Pydantic class then the model output will be a
                 Pydantic instance of that class, and the model-generated fields will be
@@ -1520,11 +1546,15 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
                 when specifying a Pydantic or `TypedDict` class.
 
             include_raw:
-                If `False` then only the parsed structured output is returned. If
-                an error occurs during model output parsing it will be raised. If `True`
-                then both the raw model response (a `BaseMessage`) and the parsed model
-                response will be returned. If an error occurs during output parsing it
-                will be caught and returned as well.
+                If `False` then only the parsed structured output is returned.
+
+                If an error occurs during model output parsing it will be raised.
+
+                If `True` then both the raw model response (a `BaseMessage`) and the
+                parsed model response will be returned.
+
+                If an error occurs during output parsing it will be caught and returned
+                as well.
 
                 The final output is always a `dict` with keys `'raw'`, `'parsed'`, and
                 `'parsing_error'`.
@@ -1602,7 +1632,7 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
         # }
         ```
 
-        Example: `dict` schema (`include_raw=False`):
+        Example: Dictionary schema (`include_raw=False`):
 
         ```python
         from pydantic import BaseModel
@@ -1629,8 +1659,9 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
         # }
         ```
 
-        !!! warning "Behavior changed in 0.2.26"
-            Added support for TypedDict class.
+        !!! warning "Behavior changed in `langchain-core` 0.2.26"
+
+            Added support for `TypedDict` class.
 
         """  # noqa: E501
         _ = kwargs.pop("method", None)
@@ -1671,41 +1702,6 @@ class BaseChatModel(BaseLanguageModel[AIMessage], ABC):
             return RunnableMap(raw=llm) | parser_with_fallback
         return llm | output_parser
 
-    @property
-    @beta()
-    def profile(self) -> ModelProfile:
-        """Return profiling information for the model.
-
-        This property will relies on the `langchain-model-profiles` package to
-        retrieve chat model capabilities, such as context window sizes and supported
-        features.
-
-        Raises:
-            ImportError: If `langchain-model-profiles` is not installed.
-
-        Returns:
-            A `ModelProfile` object containing profiling information for the model.
-        """
-        try:
-            from langchain_model_profiles import get_model_profile  # noqa: PLC0415
-        except ImportError as err:
-            informative_error_message = (
-                "To access model profiling information, please install the "
-                "`langchain-model-profiles` package: "
-                "`pip install langchain-model-profiles`."
-            )
-            raise ImportError(informative_error_message) from err
-
-        provider_id = self._llm_type
-        model_name = (
-            # Model name is not standardized across integrations. New integrations
-            # should prefer `model`.
-            getattr(self, "model", None)
-            or getattr(self, "model_name", None)
-            or getattr(self, "model_id", "")
-        )
-        return get_model_profile(provider_id, model_name) or {}
-
 
 class SimpleChatModel(BaseChatModel):
     """Simplified implementation for a chat model to inherit from.
@@ -1764,9 +1760,12 @@ def _gen_info_and_msg_metadata(
     }
 
 
+_MAX_CLEANUP_DEPTH = 100
+
+
 def _cleanup_llm_representation(serialized: Any, depth: int) -> None:
     """Remove non-serializable objects from a serialized object."""
-    if depth > 100:  # Don't cooperate for pathological cases
+    if depth > _MAX_CLEANUP_DEPTH:  # Don't cooperate for pathological cases
         return
 
     if not isinstance(serialized, dict):

libs/core/langchain_core/language_models/llms.py

@@ -651,9 +651,12 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompts: The prompts to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of the stop substrings.
-                If stop tokens are not supported consider raising NotImplementedError.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+
+                If stop tokens are not supported consider raising `NotImplementedError`.
             run_manager: Callback manager for the run.
 
         Returns:
@@ -671,9 +674,12 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompts: The prompts to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of the stop substrings.
-                If stop tokens are not supported consider raising NotImplementedError.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+
+                If stop tokens are not supported consider raising `NotImplementedError`.
             run_manager: Callback manager for the run.
 
         Returns:
@@ -705,11 +711,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompt: The prompt to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
             run_manager: Callback manager for the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Yields:
             Generation chunks.
@@ -731,11 +740,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompt: The prompt to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
             run_manager: Callback manager for the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Yields:
             Generation chunks.
@@ -846,10 +858,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompts: List of string prompts.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
             tags: List of tags to associate with each prompt. If provided, the length
                 of the list must match the length of the prompts list.
             metadata: List of metadata dictionaries to associate with each prompt. If
@@ -859,8 +875,9 @@ class BaseLLM(BaseLanguageModel[str], ABC):
                 length of the list must match the length of the prompts list.
             run_id: List of run IDs to associate with each prompt. If provided, the
                 length of the list must match the length of the prompts list.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Raises:
             ValueError: If prompts is not a list.
@@ -1116,10 +1133,14 @@ class BaseLLM(BaseLanguageModel[str], ABC):
 
         Args:
             prompts: List of string prompts.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of these substrings.
-            callbacks: `Callbacks` to pass through. Used for executing additional
-                functionality, such as logging or streaming, throughout generation.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+            callbacks: `Callbacks` to pass through.
+
+                Used for executing additional functionality, such as logging or
+                streaming, throughout generation.
             tags: List of tags to associate with each prompt. If provided, the length
                 of the list must match the length of the prompts list.
             metadata: List of metadata dictionaries to associate with each prompt. If
@@ -1129,8 +1150,9 @@ class BaseLLM(BaseLanguageModel[str], ABC):
                 length of the list must match the length of the prompts list.
             run_id: List of run IDs to associate with each prompt. If provided, the
                 length of the list must match the length of the prompts list.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Raises:
             ValueError: If the length of `callbacks`, `tags`, `metadata`, or
@@ -1410,12 +1432,16 @@ class LLM(BaseLLM):
 
         Args:
             prompt: The prompt to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of the stop substrings.
-                If stop tokens are not supported consider raising NotImplementedError.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+
+                If stop tokens are not supported consider raising `NotImplementedError`.
             run_manager: Callback manager for the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             The model output as a string. SHOULD NOT include the prompt.
@@ -1436,12 +1462,16 @@ class LLM(BaseLLM):
 
         Args:
             prompt: The prompt to generate from.
-            stop: Stop words to use when generating. Model output is cut off at the
-                first occurrence of any of the stop substrings.
-                If stop tokens are not supported consider raising NotImplementedError.
+            stop: Stop words to use when generating.
+
+                Model output is cut off at the first occurrence of any of these
+                substrings.
+
+                If stop tokens are not supported consider raising `NotImplementedError`.
             run_manager: Callback manager for the run.
-            **kwargs: Arbitrary additional keyword arguments. These are usually passed
-                to the model provider API call.
+            **kwargs: Arbitrary additional keyword arguments.
+
+                These are usually passed to the model provider API call.
 
         Returns:
             The model output as a string. SHOULD NOT include the prompt.

libs/core/langchain_core/language_models/model_profile.py

@@ -0,0 +1,84 @@
+"""Model profile types and utilities."""
+
+from typing_extensions import TypedDict
+
+
+class ModelProfile(TypedDict, total=False):
+    """Model profile.
+
+    !!! warning "Beta feature"
+        This is a beta feature. The format of model profiles is subject to change.
+
+    Provides information about chat model capabilities, such as context window sizes
+    and supported features.
+    """
+
+    # --- Input constraints ---
+
+    max_input_tokens: int
+    """Maximum context window (tokens)"""
+
+    image_inputs: bool
+    """Whether image inputs are supported."""
+    # TODO: add more detail about formats?
+
+    image_url_inputs: bool
+    """Whether [image URL inputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+
+    pdf_inputs: bool
+    """Whether [PDF inputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+    # TODO: add more detail about formats? e.g. bytes or base64
+
+    audio_inputs: bool
+    """Whether [audio inputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+    # TODO: add more detail about formats? e.g. bytes or base64
+
+    video_inputs: bool
+    """Whether [video inputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+    # TODO: add more detail about formats? e.g. bytes or base64
+
+    image_tool_message: bool
+    """Whether images can be included in tool messages."""
+
+    pdf_tool_message: bool
+    """Whether PDFs can be included in tool messages."""
+
+    # --- Output constraints ---
+
+    max_output_tokens: int
+    """Maximum output tokens"""
+
+    reasoning_output: bool
+    """Whether the model supports [reasoning / chain-of-thought](https://docs.langchain.com/oss/python/langchain/models#reasoning)"""
+
+    image_outputs: bool
+    """Whether [image outputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+
+    audio_outputs: bool
+    """Whether [audio outputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+
+    video_outputs: bool
+    """Whether [video outputs](https://docs.langchain.com/oss/python/langchain/models#multimodal)
+    are supported."""
+
+    # --- Tool calling ---
+    tool_calling: bool
+    """Whether the model supports [tool calling](https://docs.langchain.com/oss/python/langchain/models#tool-calling)"""
+
+    tool_choice: bool
+    """Whether the model supports [tool choice](https://docs.langchain.com/oss/python/langchain/models#forcing-tool-calls)"""
+
+    # --- Structured output ---
+    structured_output: bool
+    """Whether the model supports a native [structured output](https://docs.langchain.com/oss/python/langchain/models#structured-outputs)
+    feature"""
+
+
+ModelProfileRegistry = dict[str, ModelProfile]
+"""Registry mapping model identifiers or names to their ModelProfile."""

libs/core/langchain_core/load/load.py

@@ -61,13 +61,15 @@ class Reviver:
         """Initialize the reviver.
 
         Args:
-            secrets_map: A map of secrets to load. If a secret is not found in
-                the map, it will be loaded from the environment if `secrets_from_env`
-                is True.
+            secrets_map: A map of secrets to load.
+
+                If a secret is not found in the map, it will be loaded from the
+                environment if `secrets_from_env` is `True`.
             valid_namespaces: A list of additional namespaces (modules)
                 to allow to be deserialized.
             secrets_from_env: Whether to load secrets from the environment.
             additional_import_mappings: A dictionary of additional namespace mappings
+
                 You can use this to override default mappings or add new mappings.
             ignore_unserializable_fields: Whether to ignore unserializable fields.
         """
@@ -195,13 +197,15 @@ def loads(
 
     Args:
         text: The string to load.
-        secrets_map: A map of secrets to load. If a secret is not found in
-            the map, it will be loaded from the environment if `secrets_from_env`
-            is True.
+        secrets_map: A map of secrets to load.
+
+            If a secret is not found in the map, it will be loaded from the environment
+            if `secrets_from_env` is `True`.
         valid_namespaces: A list of additional namespaces (modules)
             to allow to be deserialized.
         secrets_from_env: Whether to load secrets from the environment.
         additional_import_mappings: A dictionary of additional namespace mappings
+
             You can use this to override default mappings or add new mappings.
         ignore_unserializable_fields: Whether to ignore unserializable fields.
 
@@ -237,13 +241,15 @@ def load(
 
     Args:
         obj: The object to load.
-        secrets_map: A map of secrets to load. If a secret is not found in
-            the map, it will be loaded from the environment if `secrets_from_env`
-            is True.
+        secrets_map: A map of secrets to load.
+
+            If a secret is not found in the map, it will be loaded from the environment
+            if `secrets_from_env` is `True`.
         valid_namespaces: A list of additional namespaces (modules)
             to allow to be deserialized.
         secrets_from_env: Whether to load secrets from the environment.
         additional_import_mappings: A dictionary of additional namespace mappings
+
             You can use this to override default mappings or add new mappings.
         ignore_unserializable_fields: Whether to ignore unserializable fields.
 

libs/core/langchain_core/messages/ai.py

@@ -50,7 +50,7 @@ class InputTokenDetails(TypedDict, total=False):
 
     May also hold extra provider-specific keys.
 
-    !!! version-added "Added in version 0.3.9"
+    !!! version-added "Added in `langchain-core` 0.3.9"
 
     """
 
@@ -85,7 +85,7 @@ class OutputTokenDetails(TypedDict, total=False):
 
     May also hold extra provider-specific keys.
 
-    !!! version-added "Added in version 0.3.9"
+    !!! version-added "Added in `langchain-core` 0.3.9"
 
     """
 
@@ -123,10 +123,12 @@ class UsageMetadata(TypedDict):
         }
         ```
 
-    !!! warning "Behavior changed in 0.3.9"
+    !!! warning "Behavior changed in `langchain-core` 0.3.9"
+
         Added `input_token_details` and `output_token_details`.
 
     !!! note "LangSmith SDK"
+
         The LangSmith SDK also has a `UsageMetadata` class. While the two share fields,
         LangSmith's `UsageMetadata` has additional fields to capture cost information
         used by the LangSmith platform.

libs/core/langchain_core/messages/base.py

@@ -5,11 +5,9 @@ from __future__ import annotations
 from typing import TYPE_CHECKING, Any, cast, overload
 
 from pydantic import ConfigDict, Field
-from typing_extensions import Self
 
 from langchain_core._api.deprecation import warn_deprecated
 from langchain_core.load.serializable import Serializable
-from langchain_core.messages import content as types
 from langchain_core.utils import get_bolded_text
 from langchain_core.utils._merge import merge_dicts, merge_lists
 from langchain_core.utils.interactive_env import is_interactive_env
@@ -17,6 +15,9 @@ from langchain_core.utils.interactive_env import is_interactive_env
 if TYPE_CHECKING:
     from collections.abc import Sequence
 
+    from typing_extensions import Self
+
+    from langchain_core.messages import content as types
     from langchain_core.prompts.chat import ChatPromptTemplate
 
 
@@ -199,7 +200,7 @@ class BaseMessage(Serializable):
     def content_blocks(self) -> list[types.ContentBlock]:
         r"""Load content blocks from the message content.
 
-        !!! version-added "Added in version 1.0.0"
+        !!! version-added "Added in `langchain-core` 1.0.0"
 
         """
         # Needed here to avoid circular import, as these classes import BaseMessages

libs/core/langchain_core/messages/block_translators/__init__.py

@@ -12,10 +12,11 @@ the implementation in `BaseMessage`.
 
 from __future__ import annotations
 
-from collections.abc import Callable
 from typing import TYPE_CHECKING
 
 if TYPE_CHECKING:
+    from collections.abc import Callable
+
     from langchain_core.messages import AIMessage, AIMessageChunk
     from langchain_core.messages import content as types
 

libs/core/langchain_core/messages/block_translators/openai.py

@@ -4,7 +4,6 @@ from __future__ import annotations
 
 import json
 import warnings
-from collections.abc import Iterable
 from typing import TYPE_CHECKING, Any, Literal, cast
 
 from langchain_core.language_models._utils import (
@@ -14,6 +13,8 @@ from langchain_core.language_models._utils import (
 from langchain_core.messages import content as types
 
 if TYPE_CHECKING:
+    from collections.abc import Iterable
+
     from langchain_core.messages import AIMessage, AIMessageChunk
 
 

libs/core/langchain_core/messages/content.py

@@ -654,7 +654,7 @@ class PlainTextContentBlock(TypedDict):
 
     !!! note
         Title and context are optional fields that may be passed to the model. See
-        Anthropic [example](https://docs.claude.com/en/docs/build-with-claude/citations#citable-vs-non-citable-content).
+        Anthropic [example](https://platform.claude.com/docs/en/build-with-claude/citations#citable-vs-non-citable-content).
 
     !!! note "Factory function"
         `create_plaintext_block` may also be used as a factory to create a
@@ -867,7 +867,7 @@ def _get_data_content_block_types() -> tuple[str, ...]:
     Example: ("image", "video", "audio", "text-plain", "file")
 
     Note that old style multimodal blocks type literals with new style blocks.
-    Speficially, "image", "audio", and "file".
+    Specifically, "image", "audio", and "file".
 
     See the docstring of `_normalize_messages` in `language_models._utils` for details.
     """
@@ -906,7 +906,7 @@ def is_data_content_block(block: dict) -> bool:
 
         # 'text' is checked to support v0 PlainTextContentBlock types
         # We must guard against new style TextContentBlock which also has 'text' `type`
-        # by ensuring the presense of `source_type`
+        # by ensuring the presence of `source_type`
         if block["type"] == "text" and "source_type" not in block:  # noqa: SIM103  # This is more readable
             return False
 

libs/core/langchain_core/messages/utils.py

@@ -328,12 +328,16 @@ def _convert_to_message(message: MessageLikeRepresentation) -> BaseMessage:
     """
     if isinstance(message, BaseMessage):
         message_ = message
-    elif isinstance(message, str):
-        message_ = _create_message_from_message_type("human", message)
-    elif isinstance(message, Sequence) and len(message) == 2:
-        # mypy doesn't realise this can't be a string given the previous branch
-        message_type_str, template = message  # type: ignore[misc]
-        message_ = _create_message_from_message_type(message_type_str, template)
+    elif isinstance(message, Sequence):
+        if isinstance(message, str):
+            message_ = _create_message_from_message_type("human", message)
+        else:
+            try:
+                message_type_str, template = message
+            except ValueError as e:
+                msg = "Message as a sequence must be (role string, template)"
+                raise NotImplementedError(msg) from e
+            message_ = _create_message_from_message_type(message_type_str, template)
     elif isinstance(message, dict):
         msg_kwargs = message.copy()
         try:
@@ -734,8 +738,10 @@ def trim_messages(
             Set to `len` to count the number of **messages** in the chat history.
 
             !!! note
+
                 Use `count_tokens_approximately` to get fast, approximate token
                 counts.
+
                 This is recommended for using `trim_messages` on the hot path, where
                 exact token counting is not necessary.
 
@@ -1097,7 +1103,7 @@ def convert_to_openai_messages(
         # ]
         ```
 
-    !!! version-added "Added in version 0.3.11"
+    !!! version-added "Added in `langchain-core` 0.3.11"
 
     """  # noqa: E501
     if text_format not in {"string", "block"}:
@@ -1697,7 +1703,7 @@ def count_tokens_approximately(
     Warning:
         This function does not currently support counting image tokens.
 
-    !!! version-added "Added in version 0.3.46"
+    !!! version-added "Added in `langchain-core` 0.3.46"
 
     """
     token_count = 0.0

libs/core/langchain_core/output_parsers/openai_tools.py

@@ -15,7 +15,11 @@ from langchain_core.messages.tool import tool_call as create_tool_call
 from langchain_core.output_parsers.transform import BaseCumulativeTransformOutputParser
 from langchain_core.outputs import ChatGeneration, Generation
 from langchain_core.utils.json import parse_partial_json
-from langchain_core.utils.pydantic import TypeBaseModel
+from langchain_core.utils.pydantic import (
+    TypeBaseModel,
+    is_pydantic_v1_subclass,
+    is_pydantic_v2_subclass,
+)
 
 logger = logging.getLogger(__name__)
 
@@ -323,7 +327,15 @@ class PydanticToolsParser(JsonOutputToolsParser):
             return None if self.first_tool_only else []
 
         json_results = [json_results] if self.first_tool_only else json_results
-        name_dict = {tool.__name__: tool for tool in self.tools}
+        name_dict_v2: dict[str, TypeBaseModel] = {
+            tool.model_config.get("title") or tool.__name__: tool
+            for tool in self.tools
+            if is_pydantic_v2_subclass(tool)
+        }
+        name_dict_v1: dict[str, TypeBaseModel] = {
+            tool.__name__: tool for tool in self.tools if is_pydantic_v1_subclass(tool)
+        }
+        name_dict: dict[str, TypeBaseModel] = {**name_dict_v2, **name_dict_v1}
         pydantic_objects = []
         for res in json_results:
             if not isinstance(res["args"], dict):

libs/core/langchain_core/output_parsers/pydantic.py

@@ -37,7 +37,7 @@ class PydanticOutputParser(JsonOutputParser, Generic[TBaseModel]):
     def _parser_exception(
         self, e: Exception, json_object: dict
     ) -> OutputParserException:
-        json_string = json.dumps(json_object)
+        json_string = json.dumps(json_object, ensure_ascii=False)
         name = self.pydantic_object.__name__
         msg = f"Failed to parse {name} from completion {json_string}. Got: {e}"
         return OutputParserException(msg, llm_output=json_string)

libs/core/langchain_core/outputs/chat_generation.py

@@ -2,15 +2,17 @@
 
 from __future__ import annotations
 
-from typing import Literal
+from typing import TYPE_CHECKING, Literal
 
 from pydantic import model_validator
-from typing_extensions import Self
 
 from langchain_core.messages import BaseMessage, BaseMessageChunk
 from langchain_core.outputs.generation import Generation
 from langchain_core.utils._merge import merge_dicts
 
+if TYPE_CHECKING:
+    from typing_extensions import Self
+
 
 class ChatGeneration(Generation):
     """A single chat generation output.

libs/core/langchain_core/outputs/generation.py

@@ -20,8 +20,7 @@ class Generation(Serializable):
 
     LangChain users working with chat models will usually access information via
     `AIMessage` (returned from runnable interfaces) or `LLMResult` (available
-    via callbacks). Please refer the `AIMessage` and `LLMResult` schema documentation
-    for more information.
+    via callbacks). Please refer to `AIMessage` and `LLMResult` for more information.
     """
 
     text: str
@@ -34,11 +33,13 @@ class Generation(Serializable):
     """
     type: Literal["Generation"] = "Generation"
     """Type is used exclusively for serialization purposes.
-    Set to "Generation" for this class."""
+
+    Set to "Generation" for this class.
+    """
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -52,7 +53,7 @@ class Generation(Serializable):
 
 
 class GenerationChunk(Generation):
-    """Generation chunk, which can be concatenated with other Generation chunks."""
+    """`GenerationChunk`, which can be concatenated with other Generation chunks."""
 
     def __add__(self, other: GenerationChunk) -> GenerationChunk:
         """Concatenate two `GenerationChunk`s.

libs/core/langchain_core/prompt_values.py

@@ -30,7 +30,7 @@ class PromptValue(Serializable, ABC):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -48,7 +48,7 @@ class PromptValue(Serializable, ABC):
 
     @abstractmethod
     def to_messages(self) -> list[BaseMessage]:
-        """Return prompt as a list of Messages."""
+        """Return prompt as a list of messages."""
 
 
 class StringPromptValue(PromptValue):

libs/core/langchain_core/prompts/base.py

@@ -6,7 +6,7 @@ import contextlib
 import json
 import typing
 from abc import ABC, abstractmethod
-from collections.abc import Callable, Mapping
+from collections.abc import Mapping
 from functools import cached_property
 from pathlib import Path
 from typing import (
@@ -33,6 +33,8 @@ from langchain_core.runnables.config import ensure_config
 from langchain_core.utils.pydantic import create_model_v2
 
 if TYPE_CHECKING:
+    from collections.abc import Callable
+
     from langchain_core.documents import Document
 
 
@@ -46,23 +48,27 @@ class BasePromptTemplate(
 
     input_variables: list[str]
     """A list of the names of the variables whose values are required as inputs to the
-    prompt."""
+    prompt.
+    """
     optional_variables: list[str] = Field(default=[])
     """A list of the names of the variables for placeholder or `MessagePlaceholder` that
     are optional.
 
-    These variables are auto inferred from the prompt and user need not provide them."""
+    These variables are auto inferred from the prompt and user need not provide them.
+    """
     input_types: typing.Dict[str, Any] = Field(default_factory=dict, exclude=True)  # noqa: UP006
     """A dictionary of the types of the variables the prompt template expects.
 
-    If not provided, all variables are assumed to be strings."""
+    If not provided, all variables are assumed to be strings.
+    """
     output_parser: BaseOutputParser | None = None
     """How to parse the output of calling an LLM on this formatted prompt."""
     partial_variables: Mapping[str, Any] = Field(default_factory=dict)
     """A dictionary of the partial variables the prompt template carries.
 
-    Partial variables populate the template so that you don't need to
-    pass them in every time you call the prompt."""
+    Partial variables populate the template so that you don't need to pass them in every
+    time you call the prompt.
+    """
     metadata: typing.Dict[str, Any] | None = None  # noqa: UP006
     """Metadata to be used for tracing."""
     tags: list[str] | None = None
@@ -107,7 +113,7 @@ class BasePromptTemplate(
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     model_config = ConfigDict(
@@ -129,7 +135,7 @@ class BasePromptTemplate(
         """Get the input schema for the prompt.
 
         Args:
-            config: configuration for the prompt.
+            config: Configuration for the prompt.
 
         Returns:
             The input schema for the prompt.
@@ -197,8 +203,8 @@ class BasePromptTemplate(
         """Invoke the prompt.
 
         Args:
-            input: Dict, input to the prompt.
-            config: RunnableConfig, configuration for the prompt.
+            input: Input to the prompt.
+            config: Configuration for the prompt.
 
         Returns:
             The output of the prompt.
@@ -223,8 +229,8 @@ class BasePromptTemplate(
         """Async invoke the prompt.
 
         Args:
-            input: Dict, input to the prompt.
-            config: RunnableConfig, configuration for the prompt.
+            input: Input to the prompt.
+            config: Configuration for the prompt.
 
         Returns:
             The output of the prompt.
@@ -244,7 +250,7 @@ class BasePromptTemplate(
 
     @abstractmethod
     def format_prompt(self, **kwargs: Any) -> PromptValue:
-        """Create Prompt Value.
+        """Create `PromptValue`.
 
         Args:
             **kwargs: Any arguments to be passed to the prompt template.
@@ -254,7 +260,7 @@ class BasePromptTemplate(
         """
 
     async def aformat_prompt(self, **kwargs: Any) -> PromptValue:
-        """Async create Prompt Value.
+        """Async create `PromptValue`.
 
         Args:
             **kwargs: Any arguments to be passed to the prompt template.
@@ -268,7 +274,7 @@ class BasePromptTemplate(
         """Return a partial of the prompt template.
 
         Args:
-            **kwargs: partial variables to set.
+            **kwargs: Partial variables to set.
 
         Returns:
             A partial of the prompt template.
@@ -298,9 +304,9 @@ class BasePromptTemplate(
             A formatted string.
 
         Example:
-        ```python
-        prompt.format(variable1="foo")
-        ```
+            ```python
+            prompt.format(variable1="foo")
+            ```
         """
 
     async def aformat(self, **kwargs: Any) -> FormatOutputType:
@@ -313,9 +319,9 @@ class BasePromptTemplate(
             A formatted string.
 
         Example:
-        ```python
-        await prompt.aformat(variable1="foo")
-        ```
+            ```python
+            await prompt.aformat(variable1="foo")
+            ```
         """
         return self.format(**kwargs)
 
@@ -350,9 +356,9 @@ class BasePromptTemplate(
             NotImplementedError: If the prompt type is not implemented.
 
         Example:
-        ```python
-        prompt.save(file_path="path/prompt.yaml")
-        ```
+            ```python
+            prompt.save(file_path="path/prompt.yaml")
+            ```
         """
         if self.partial_variables:
             msg = "Cannot save prompt with partial variables."
@@ -404,23 +410,23 @@ def format_document(doc: Document, prompt: BasePromptTemplate[str]) -> str:
 
     First, this pulls information from the document from two sources:
 
-    1. page_content:
-        This takes the information from the `document.page_content`
-        and assigns it to a variable named `page_content`.
-    2. metadata:
-        This takes information from `document.metadata` and assigns
-        it to variables of the same name.
+    1. `page_content`:
+        This takes the information from the `document.page_content` and assigns it to a
+        variable named `page_content`.
+    2. `metadata`:
+        This takes information from `document.metadata` and assigns it to variables of
+        the same name.
 
     Those variables are then passed into the `prompt` to produce a formatted string.
 
     Args:
-        doc: Document, the page_content and metadata will be used to create
+        doc: `Document`, the `page_content` and `metadata` will be used to create
             the final string.
-        prompt: BasePromptTemplate, will be used to format the page_content
-            and metadata into the final string.
+        prompt: `BasePromptTemplate`, will be used to format the `page_content`
+            and `metadata` into the final string.
 
     Returns:
-        string of the document formatted.
+        String of the document formatted.
 
     Example:
         ```python
@@ -431,7 +437,6 @@ def format_document(doc: Document, prompt: BasePromptTemplate[str]) -> str:
         prompt = PromptTemplate.from_template("Page {page}: {page_content}")
         format_document(doc, prompt)
         >>> "Page 1: This is a joke"
-
         ```
     """
     return prompt.format(**_get_document_info(doc, prompt))
@@ -442,22 +447,22 @@ async def aformat_document(doc: Document, prompt: BasePromptTemplate[str]) -> st
 
     First, this pulls information from the document from two sources:
 
-    1. page_content:
-        This takes the information from the `document.page_content`
-        and assigns it to a variable named `page_content`.
-    2. metadata:
-        This takes information from `document.metadata` and assigns
-        it to variables of the same name.
+    1. `page_content`:
+        This takes the information from the `document.page_content` and assigns it to a
+        variable named `page_content`.
+    2. `metadata`:
+        This takes information from `document.metadata` and assigns it to variables of
+        the same name.
 
     Those variables are then passed into the `prompt` to produce a formatted string.
 
     Args:
-        doc: Document, the page_content and metadata will be used to create
+        doc: `Document`, the `page_content` and `metadata` will be used to create
             the final string.
-        prompt: BasePromptTemplate, will be used to format the page_content
-            and metadata into the final string.
+        prompt: `BasePromptTemplate`, will be used to format the `page_content`
+            and `metadata` into the final string.
 
     Returns:
-        string of the document formatted.
+        String of the document formatted.
     """
     return await prompt.aformat(**_get_document_info(doc, prompt))

libs/core/langchain_core/prompts/chat.py

@@ -587,14 +587,15 @@ class _StringImageMessagePromptTemplate(BaseMessagePromptTemplate):
         for prompt in self.prompt:
             inputs = {var: kwargs[var] for var in prompt.input_variables}
             if isinstance(prompt, StringPromptTemplate):
-                formatted: str | ImageURL | dict[str, Any] = prompt.format(**inputs)
-                content.append({"type": "text", "text": formatted})
+                formatted_text: str = prompt.format(**inputs)
+                if formatted_text != "":
+                    content.append({"type": "text", "text": formatted_text})
             elif isinstance(prompt, ImagePromptTemplate):
-                formatted = prompt.format(**inputs)
-                content.append({"type": "image_url", "image_url": formatted})
+                formatted_image: ImageURL = prompt.format(**inputs)
+                content.append({"type": "image_url", "image_url": formatted_image})
             elif isinstance(prompt, DictPromptTemplate):
-                formatted = prompt.format(**inputs)
-                content.append(formatted)
+                formatted_dict: dict[str, Any] = prompt.format(**inputs)
+                content.append(formatted_dict)
         return self._msg_class(
             content=content, additional_kwargs=self.additional_kwargs
         )
@@ -617,16 +618,15 @@ class _StringImageMessagePromptTemplate(BaseMessagePromptTemplate):
         for prompt in self.prompt:
             inputs = {var: kwargs[var] for var in prompt.input_variables}
             if isinstance(prompt, StringPromptTemplate):
-                formatted: str | ImageURL | dict[str, Any] = await prompt.aformat(
-                    **inputs
-                )
-                content.append({"type": "text", "text": formatted})
+                formatted_text: str = await prompt.aformat(**inputs)
+                if formatted_text != "":
+                    content.append({"type": "text", "text": formatted_text})
             elif isinstance(prompt, ImagePromptTemplate):
-                formatted = await prompt.aformat(**inputs)
-                content.append({"type": "image_url", "image_url": formatted})
+                formatted_image: ImageURL = await prompt.aformat(**inputs)
+                content.append({"type": "image_url", "image_url": formatted_image})
             elif isinstance(prompt, DictPromptTemplate):
-                formatted = prompt.format(**inputs)
-                content.append(formatted)
+                formatted_dict: dict[str, Any] = prompt.format(**inputs)
+                content.append(formatted_dict)
         return self._msg_class(
             content=content, additional_kwargs=self.additional_kwargs
         )
@@ -903,23 +903,28 @@ class ChatPromptTemplate(BaseChatPromptTemplate):
                 5. A string which is shorthand for `("human", template)`; e.g.,
                     `"{user_input}"`
             template_format: Format of the template.
-            input_variables: A list of the names of the variables whose values are
-                required as inputs to the prompt.
-            optional_variables: A list of the names of the variables for placeholder
-                or MessagePlaceholder that are optional.
+            **kwargs: Additional keyword arguments passed to `BasePromptTemplate`,
+                including (but not limited to):
 
-                These variables are auto inferred from the prompt and user need not
-                provide them.
-            partial_variables: A dictionary of the partial variables the prompt
-                template carries.
+                - `input_variables`: A list of the names of the variables whose values
+                    are required as inputs to the prompt.
+                - `optional_variables`: A list of the names of the variables for
+                    placeholder or `MessagePlaceholder` that are optional.
 
-                Partial variables populate the template so that you don't need to pass
-                them in every time you call the prompt.
-            validate_template: Whether to validate the template.
-            input_types: A dictionary of the types of the variables the prompt template
-                expects.
+                    These variables are auto inferred from the prompt and user need not
+                    provide them.
 
-                If not provided, all variables are assumed to be strings.
+                - `partial_variables`: A dictionary of the partial variables the prompt
+                    template carries.
+
+                    Partial variables populate the template so that you don't need to
+                    pass them in every time you call the prompt.
+
+                - `validate_template`: Whether to validate the template.
+                - `input_types`: A dictionary of the types of the variables the prompt
+                    template expects.
+
+                    If not provided, all variables are assumed to be strings.
 
         Examples:
             Instantiation from a list of message templates:
@@ -1343,11 +1348,25 @@ def _create_template_from_message_type(
                 raise ValueError(msg)
             var_name = template[1:-1]
             message = MessagesPlaceholder(variable_name=var_name, optional=True)
-        elif len(template) == 2 and isinstance(template[1], bool):
-            var_name_wrapped, is_optional = template
+        else:
+            try:
+                var_name_wrapped, is_optional = template
+            except ValueError as e:
+                msg = (
+                    "Unexpected arguments for placeholder message type."
+                    " Expected either a single string variable name"
+                    " or a list of [variable_name: str, is_optional: bool]."
+                    f" Got: {template}"
+                )
+                raise ValueError(msg) from e
+
+            if not isinstance(is_optional, bool):
+                msg = f"Expected is_optional to be a boolean. Got: {is_optional}"
+                raise ValueError(msg)  # noqa: TRY004
+
             if not isinstance(var_name_wrapped, str):
                 msg = f"Expected variable name to be a string. Got: {var_name_wrapped}"
-                raise ValueError(msg)  # noqa:TRY004
+                raise ValueError(msg)  # noqa: TRY004
             if var_name_wrapped[0] != "{" or var_name_wrapped[-1] != "}":
                 msg = (
                     f"Invalid placeholder template: {var_name_wrapped}."
@@ -1357,14 +1376,6 @@ def _create_template_from_message_type(
             var_name = var_name_wrapped[1:-1]
 
             message = MessagesPlaceholder(variable_name=var_name, optional=is_optional)
-        else:
-            msg = (
-                "Unexpected arguments for placeholder message type."
-                " Expected either a single string variable name"
-                " or a list of [variable_name: str, is_optional: bool]."
-                f" Got: {template}"
-            )
-            raise ValueError(msg)
     else:
         msg = (
             f"Unexpected message type: {message_type}. Use one of 'human',"
@@ -1418,10 +1429,11 @@ def _convert_to_message_template(
                 )
                 raise ValueError(msg)
             message = (message["role"], message["content"])
-        if len(message) != 2:
+        try:
+            message_type_str, template = message
+        except ValueError as e:
             msg = f"Expected 2-tuple of (role, template), got {message}"
-            raise ValueError(msg)
-        message_type_str, template = message
+            raise ValueError(msg) from e
         if isinstance(message_type_str, str):
             message_ = _create_template_from_message_type(
                 message_type_str, template, template_format=template_format

libs/core/langchain_core/prompts/dict.py

@@ -69,7 +69,7 @@ class DictPromptTemplate(RunnableSerializable[dict, dict]):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod

libs/core/langchain_core/prompts/message.py

@@ -6,10 +6,10 @@ from abc import ABC, abstractmethod
 from typing import TYPE_CHECKING, Any
 
 from langchain_core.load import Serializable
-from langchain_core.messages import BaseMessage
 from langchain_core.utils.interactive_env import is_interactive_env
 
 if TYPE_CHECKING:
+    from langchain_core.messages import BaseMessage
     from langchain_core.prompts.chat import ChatPromptTemplate
 
 
@@ -18,7 +18,7 @@ class BaseMessagePromptTemplate(Serializable, ABC):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -32,13 +32,13 @@ class BaseMessagePromptTemplate(Serializable, ABC):
 
     @abstractmethod
     def format_messages(self, **kwargs: Any) -> list[BaseMessage]:
-        """Format messages from kwargs. Should return a list of BaseMessages.
+        """Format messages from kwargs. Should return a list of `BaseMessage` objects.
 
         Args:
             **kwargs: Keyword arguments to use for formatting.
 
         Returns:
-            List of BaseMessages.
+            List of `BaseMessage` objects.
         """
 
     async def aformat_messages(self, **kwargs: Any) -> list[BaseMessage]:
@@ -48,7 +48,7 @@ class BaseMessagePromptTemplate(Serializable, ABC):
             **kwargs: Keyword arguments to use for formatting.
 
         Returns:
-            List of BaseMessages.
+            List of `BaseMessage` objects.
         """
         return self.format_messages(**kwargs)
 

libs/core/langchain_core/prompts/string.py

@@ -4,9 +4,8 @@ from __future__ import annotations
 
 import warnings
 from abc import ABC
-from collections.abc import Callable, Sequence
 from string import Formatter
-from typing import Any, Literal
+from typing import TYPE_CHECKING, Any, Literal
 
 from pydantic import BaseModel, create_model
 
@@ -16,10 +15,70 @@ from langchain_core.utils import get_colored_text, mustache
 from langchain_core.utils.formatting import formatter
 from langchain_core.utils.interactive_env import is_interactive_env
 
+if TYPE_CHECKING:
+    from collections.abc import Callable, Sequence
+
 try:
-    from jinja2 import Environment, meta
+    from jinja2 import meta
+    from jinja2.exceptions import SecurityError
     from jinja2.sandbox import SandboxedEnvironment
 
+    class _RestrictedSandboxedEnvironment(SandboxedEnvironment):
+        """A more restrictive Jinja2 sandbox that blocks all attribute/method access.
+
+        This sandbox only allows simple variable lookups, no attribute or method access.
+        This prevents template injection attacks via methods like parse_raw().
+        """
+
+        def is_safe_attribute(self, _obj: Any, _attr: str, _value: Any) -> bool:
+            """Block ALL attribute access for security.
+
+            Only allow accessing variables directly from the context dict,
+            no attribute access on those objects.
+
+            Args:
+                _obj: The object being accessed (unused, always blocked).
+                _attr: The attribute name (unused, always blocked).
+                _value: The attribute value (unused, always blocked).
+
+            Returns:
+                False - all attribute access is blocked.
+            """
+            # Block all attribute access
+            return False
+
+        def is_safe_callable(self, _obj: Any) -> bool:
+            """Block all method calls for security.
+
+            Args:
+                _obj: The object being checked (unused, always blocked).
+
+            Returns:
+                False - all callables are blocked.
+            """
+            return False
+
+        def getattr(self, obj: Any, attribute: str) -> Any:
+            """Override getattr to block all attribute access.
+
+            Args:
+                obj: The object.
+                attribute: The attribute name.
+
+            Returns:
+                Never returns.
+
+            Raises:
+                SecurityError: Always, to block attribute access.
+            """
+            msg = (
+                f"Access to attributes is not allowed in templates. "
+                f"Attempted to access '{attribute}' on {type(obj).__name__}. "
+                f"Use only simple variable names like {{{{variable}}}} "
+                f"without dots or methods."
+            )
+            raise SecurityError(msg)
+
     _HAS_JINJA2 = True
 except ImportError:
     _HAS_JINJA2 = False
@@ -59,14 +118,10 @@ def jinja2_formatter(template: str, /, **kwargs: Any) -> str:
         )
         raise ImportError(msg)
 
-    # This uses a sandboxed environment to prevent arbitrary code execution.
-    # Jinja2 uses an opt-out rather than opt-in approach for sand-boxing.
-    # Please treat this sand-boxing as a best-effort approach rather than
-    # a guarantee of security.
-    # We recommend to never use jinja2 templates with untrusted inputs.
-    # https://jinja.palletsprojects.com/en/3.1.x/sandbox/
-    # approach not a guarantee of security.
-    return SandboxedEnvironment().from_string(template).render(**kwargs)
+    # Use a restricted sandbox that blocks ALL attribute/method access
+    # Only simple variable lookups like {{variable}} are allowed
+    # Attribute access like {{variable.attr}} or {{variable.method()}} is blocked
+    return _RestrictedSandboxedEnvironment().from_string(template).render(**kwargs)
 
 
 def validate_jinja2(template: str, input_variables: list[str]) -> None:
@@ -101,7 +156,7 @@ def _get_jinja2_variables_from_template(template: str) -> set[str]:
             "Please install it with `pip install jinja2`."
         )
         raise ImportError(msg)
-    env = Environment()  # noqa: S701
+    env = _RestrictedSandboxedEnvironment()
     ast = env.parse(template)
     return meta.find_undeclared_variables(ast)
 
@@ -271,6 +326,30 @@ def get_template_variables(template: str, template_format: str) -> list[str]:
         msg = f"Unsupported template format: {template_format}"
         raise ValueError(msg)
 
+    # For f-strings, block attribute access and indexing syntax
+    # This prevents template injection attacks via accessing dangerous attributes
+    if template_format == "f-string":
+        for var in input_variables:
+            # Formatter().parse() returns field names with dots/brackets if present
+            # e.g., "obj.attr" or "obj[0]" - we need to block these
+            if "." in var or "[" in var or "]" in var:
+                msg = (
+                    f"Invalid variable name {var!r} in f-string template. "
+                    f"Variable names cannot contain attribute "
+                    f"access (.) or indexing ([])."
+                )
+                raise ValueError(msg)
+
+            # Block variable names that are all digits (e.g., "0", "100")
+            # These are interpreted as positional arguments, not keyword arguments
+            if var.isdigit():
+                msg = (
+                    f"Invalid variable name {var!r} in f-string template. "
+                    f"Variable names cannot be all digits as they are interpreted "
+                    f"as positional arguments."
+                )
+                raise ValueError(msg)
+
     return sorted(input_variables)
 
 

libs/core/langchain_core/prompts/structured.py

@@ -49,7 +49,13 @@ class StructuredPrompt(ChatPromptTemplate):
             structured_output_kwargs: additional kwargs for structured output.
             template_format: template format for the prompt.
         """
-        schema_ = schema_ or kwargs.pop("schema")
+        schema_ = schema_ or kwargs.pop("schema", None)
+        if not schema_:
+            err_msg = (
+                "Must pass in a non-empty structured output schema. Received: "
+                f"{schema_}"
+            )
+            raise ValueError(err_msg)
         structured_output_kwargs = structured_output_kwargs or {}
         for k in set(kwargs).difference(get_pydantic_field_names(self.__class__)):
             structured_output_kwargs[k] = kwargs.pop(k)

libs/core/langchain_core/runnables/base.py

@@ -118,6 +118,8 @@ if TYPE_CHECKING:
 
 Other = TypeVar("Other")
 
+_RUNNABLE_GENERIC_NUM_ARGS = 2  # Input and Output
+
 
 class Runnable(ABC, Generic[Input, Output]):
     """A unit of work that can be invoked, batched, streamed, transformed and composed.
@@ -309,7 +311,10 @@ class Runnable(ABC, Generic[Input, Output]):
         for base in self.__class__.mro():
             if hasattr(base, "__pydantic_generic_metadata__"):
                 metadata = base.__pydantic_generic_metadata__
-                if "args" in metadata and len(metadata["args"]) == 2:
+                if (
+                    "args" in metadata
+                    and len(metadata["args"]) == _RUNNABLE_GENERIC_NUM_ARGS
+                ):
                     return metadata["args"][0]
 
         # If we didn't find a Pydantic model in the parent classes,
@@ -317,7 +322,7 @@ class Runnable(ABC, Generic[Input, Output]):
         # Runnables that are not pydantic models.
         for cls in self.__class__.__orig_bases__:  # type: ignore[attr-defined]
             type_args = get_args(cls)
-            if type_args and len(type_args) == 2:
+            if type_args and len(type_args) == _RUNNABLE_GENERIC_NUM_ARGS:
                 return type_args[0]
 
         msg = (
@@ -340,12 +345,15 @@ class Runnable(ABC, Generic[Input, Output]):
         for base in self.__class__.mro():
             if hasattr(base, "__pydantic_generic_metadata__"):
                 metadata = base.__pydantic_generic_metadata__
-                if "args" in metadata and len(metadata["args"]) == 2:
+                if (
+                    "args" in metadata
+                    and len(metadata["args"]) == _RUNNABLE_GENERIC_NUM_ARGS
+                ):
                     return metadata["args"][1]
 
         for cls in self.__class__.__orig_bases__:  # type: ignore[attr-defined]
             type_args = get_args(cls)
-            if type_args and len(type_args) == 2:
+            if type_args and len(type_args) == _RUNNABLE_GENERIC_NUM_ARGS:
                 return type_args[1]
 
         msg = (
@@ -424,7 +432,7 @@ class Runnable(ABC, Generic[Input, Output]):
             print(runnable.get_input_jsonschema())
             ```
 
-        !!! version-added "Added in version 0.3.0"
+        !!! version-added "Added in `langchain-core` 0.3.0"
 
         """
         return self.get_input_schema(config).model_json_schema()
@@ -502,7 +510,7 @@ class Runnable(ABC, Generic[Input, Output]):
             print(runnable.get_output_jsonschema())
             ```
 
-        !!! version-added "Added in version 0.3.0"
+        !!! version-added "Added in `langchain-core` 0.3.0"
 
         """
         return self.get_output_schema(config).model_json_schema()
@@ -566,7 +574,7 @@ class Runnable(ABC, Generic[Input, Output]):
         Returns:
             A JSON schema that represents the config of the `Runnable`.
 
-        !!! version-added "Added in version 0.3.0"
+        !!! version-added "Added in `langchain-core` 0.3.0"
 
         """
         return self.config_schema(include=include).model_json_schema()
@@ -699,51 +707,53 @@ class Runnable(ABC, Generic[Input, Output]):
     def pick(self, keys: str | list[str]) -> RunnableSerializable[Any, Any]:
         """Pick keys from the output `dict` of this `Runnable`.
 
-        Pick a single key:
+        !!! example "Pick a single key"
 
-        ```python
-        import json
+            ```python
+            import json
 
-        from langchain_core.runnables import RunnableLambda, RunnableMap
+            from langchain_core.runnables import RunnableLambda, RunnableMap
 
-        as_str = RunnableLambda(str)
-        as_json = RunnableLambda(json.loads)
-        chain = RunnableMap(str=as_str, json=as_json)
+            as_str = RunnableLambda(str)
+            as_json = RunnableLambda(json.loads)
+            chain = RunnableMap(str=as_str, json=as_json)
 
-        chain.invoke("[1, 2, 3]")
-        # -> {"str": "[1, 2, 3]", "json": [1, 2, 3]}
+            chain.invoke("[1, 2, 3]")
+            # -> {"str": "[1, 2, 3]", "json": [1, 2, 3]}
 
-        json_only_chain = chain.pick("json")
-        json_only_chain.invoke("[1, 2, 3]")
-        # -> [1, 2, 3]
-        ```
+            json_only_chain = chain.pick("json")
+            json_only_chain.invoke("[1, 2, 3]")
+            # -> [1, 2, 3]
+            ```
 
-        Pick a list of keys:
+        !!! example "Pick a list of keys"
 
-        ```python
-        from typing import Any
+            ```python
+            from typing import Any
 
-        import json
+            import json
 
-        from langchain_core.runnables import RunnableLambda, RunnableMap
+            from langchain_core.runnables import RunnableLambda, RunnableMap
 
-        as_str = RunnableLambda(str)
-        as_json = RunnableLambda(json.loads)
+            as_str = RunnableLambda(str)
+            as_json = RunnableLambda(json.loads)
 
 
-        def as_bytes(x: Any) -> bytes:
-            return bytes(x, "utf-8")
+            def as_bytes(x: Any) -> bytes:
+                return bytes(x, "utf-8")
 
 
-        chain = RunnableMap(str=as_str, json=as_json, bytes=RunnableLambda(as_bytes))
+            chain = RunnableMap(
+                str=as_str, json=as_json, bytes=RunnableLambda(as_bytes)
+            )
 
-        chain.invoke("[1, 2, 3]")
-        # -> {"str": "[1, 2, 3]", "json": [1, 2, 3], "bytes": b"[1, 2, 3]"}
+            chain.invoke("[1, 2, 3]")
+            # -> {"str": "[1, 2, 3]", "json": [1, 2, 3], "bytes": b"[1, 2, 3]"}
 
-        json_and_bytes_chain = chain.pick(["json", "bytes"])
-        json_and_bytes_chain.invoke("[1, 2, 3]")
-        # -> {"json": [1, 2, 3], "bytes": b"[1, 2, 3]"}
-        ```
+            json_and_bytes_chain = chain.pick(["json", "bytes"])
+            json_and_bytes_chain.invoke("[1, 2, 3]")
+            # -> {"json": [1, 2, 3], "bytes": b"[1, 2, 3]"}
+            ```
 
         Args:
             keys: A key or list of keys to pick from the output dict.
@@ -766,7 +776,7 @@ class Runnable(ABC, Generic[Input, Output]):
         """Assigns new fields to the `dict` output of this `Runnable`.
 
         ```python
-        from langchain_community.llms.fake import FakeStreamingListLLM
+        from langchain_core.language_models.fake import FakeStreamingListLLM
         from langchain_core.output_parsers import StrOutputParser
         from langchain_core.prompts import SystemMessagePromptTemplate
         from langchain_core.runnables import Runnable
@@ -818,10 +828,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             input: The input to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
 
         Returns:
             The output of the `Runnable`.
@@ -838,10 +850,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             input: The input to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
 
         Returns:
             The output of the `Runnable`.
@@ -868,8 +882,9 @@ class Runnable(ABC, Generic[Input, Output]):
             config: A config to use when invoking the `Runnable`. The config supports
                 standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work
-                to do in parallel, and other keys. Please refer to the
-                `RunnableConfig` for more details.
+                to do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
             return_exceptions: Whether to return exceptions instead of raising them.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
@@ -932,10 +947,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             inputs: A list of inputs to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
             return_exceptions: Whether to return exceptions instead of raising them.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
@@ -998,10 +1015,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             inputs: A list of inputs to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
             return_exceptions: Whether to return exceptions instead of raising them.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
@@ -1061,10 +1080,12 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             inputs: A list of inputs to the `Runnable`.
             config: A config to use when invoking the `Runnable`.
+
                 The config supports standard keys like `'tags'`, `'metadata'` for
                 tracing purposes, `'max_concurrency'` for controlling how much work to
-                do in parallel, and other keys. Please refer to the `RunnableConfig`
-                for more details.
+                do in parallel, and other keys.
+
+                Please refer to `RunnableConfig` for more details.
             return_exceptions: Whether to return exceptions instead of raising them.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
@@ -1353,48 +1374,50 @@ class Runnable(ABC, Generic[Input, Output]):
         ).with_config({"run_name": "my_template", "tags": ["my_template"]})
         ```
 
-        For instance:
+        !!! example
 
-        ```python
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from langchain_core.runnables import RunnableLambda
 
 
-        async def reverse(s: str) -> str:
-            return s[::-1]
+            async def reverse(s: str) -> str:
+                return s[::-1]
 
 
-        chain = RunnableLambda(func=reverse)
+            chain = RunnableLambda(func=reverse)
 
-        events = [event async for event in chain.astream_events("hello", version="v2")]
+            events = [
+                event async for event in chain.astream_events("hello", version="v2")
+            ]
 
-        # Will produce the following events
-        # (run_id, and parent_ids has been omitted for brevity):
-        [
-            {
-                "data": {"input": "hello"},
-                "event": "on_chain_start",
-                "metadata": {},
-                "name": "reverse",
-                "tags": [],
-            },
-            {
-                "data": {"chunk": "olleh"},
-                "event": "on_chain_stream",
-                "metadata": {},
-                "name": "reverse",
-                "tags": [],
-            },
-            {
-                "data": {"output": "olleh"},
-                "event": "on_chain_end",
-                "metadata": {},
-                "name": "reverse",
-                "tags": [],
-            },
-        ]
-        ```
+            # Will produce the following events
+            # (run_id, and parent_ids has been omitted for brevity):
+            [
+                {
+                    "data": {"input": "hello"},
+                    "event": "on_chain_start",
+                    "metadata": {},
+                    "name": "reverse",
+                    "tags": [],
+                },
+                {
+                    "data": {"chunk": "olleh"},
+                    "event": "on_chain_stream",
+                    "metadata": {},
+                    "name": "reverse",
+                    "tags": [],
+                },
+                {
+                    "data": {"output": "olleh"},
+                    "event": "on_chain_end",
+                    "metadata": {},
+                    "name": "reverse",
+                    "tags": [],
+                },
+            ]
+            ```
 
-        ```python title="Example: Dispatch Custom Event"
+        ```python title="Dispatch custom event"
         from langchain_core.callbacks.manager import (
             adispatch_custom_event,
         )
@@ -1428,10 +1451,13 @@ class Runnable(ABC, Generic[Input, Output]):
         Args:
             input: The input to the `Runnable`.
             config: The config to use for the `Runnable`.
-            version: The version of the schema to use either `'v2'` or `'v1'`.
+            version: The version of the schema to use, either `'v2'` or `'v1'`.
+
                 Users should use `'v2'`.
+
                 `'v1'` is for backwards compatibility and will be deprecated
                 in `0.4.0`.
+
                 No default will be assigned until the API is stabilized.
                 custom events will only be surfaced in `'v2'`.
             include_names: Only include events from `Runnable` objects with matching names.
@@ -1441,6 +1467,7 @@ class Runnable(ABC, Generic[Input, Output]):
             exclude_types: Exclude events from `Runnable` objects with matching types.
             exclude_tags: Exclude events from `Runnable` objects with matching tags.
             **kwargs: Additional keyword arguments to pass to the `Runnable`.
+
                 These will be passed to `astream_log` as this implementation
                 of `astream_events` is built on top of `astream_log`.
 
@@ -1742,46 +1769,52 @@ class Runnable(ABC, Generic[Input, Output]):
             import time
             import asyncio
 
+
             def format_t(timestamp: float) -> str:
                 return datetime.fromtimestamp(timestamp, tz=timezone.utc).isoformat()
 
+
             async def test_runnable(time_to_sleep: int):
                 print(f"Runnable[{time_to_sleep}s]: starts at {format_t(time.time())}")
                 await asyncio.sleep(time_to_sleep)
                 print(f"Runnable[{time_to_sleep}s]: ends at {format_t(time.time())}")
 
+
             async def fn_start(run_obj: Runnable):
                 print(f"on start callback starts at {format_t(time.time())}")
                 await asyncio.sleep(3)
                 print(f"on start callback ends at {format_t(time.time())}")
 
+
             async def fn_end(run_obj: Runnable):
                 print(f"on end callback starts at {format_t(time.time())}")
                 await asyncio.sleep(2)
                 print(f"on end callback ends at {format_t(time.time())}")
 
+
             runnable = RunnableLambda(test_runnable).with_alisteners(
-                on_start=fn_start,
-                on_end=fn_end
+                on_start=fn_start, on_end=fn_end
             )
+
+
             async def concurrent_runs():
                 await asyncio.gather(runnable.ainvoke(2), runnable.ainvoke(3))
 
-            asyncio.run(concurrent_runs())
-            Result:
-            on start callback starts at 2025-03-01T07:05:22.875378+00:00
-            on start callback starts at 2025-03-01T07:05:22.875495+00:00
-            on start callback ends at 2025-03-01T07:05:25.878862+00:00
-            on start callback ends at 2025-03-01T07:05:25.878947+00:00
-            Runnable[2s]: starts at 2025-03-01T07:05:25.879392+00:00
-            Runnable[3s]: starts at 2025-03-01T07:05:25.879804+00:00
-            Runnable[2s]: ends at 2025-03-01T07:05:27.881998+00:00
-            on end callback starts at 2025-03-01T07:05:27.882360+00:00
-            Runnable[3s]: ends at 2025-03-01T07:05:28.881737+00:00
-            on end callback starts at 2025-03-01T07:05:28.882428+00:00
-            on end callback ends at 2025-03-01T07:05:29.883893+00:00
-            on end callback ends at 2025-03-01T07:05:30.884831+00:00
 
+            asyncio.run(concurrent_runs())
+            # Result:
+            # on start callback starts at 2025-03-01T07:05:22.875378+00:00
+            # on start callback starts at 2025-03-01T07:05:22.875495+00:00
+            # on start callback ends at 2025-03-01T07:05:25.878862+00:00
+            # on start callback ends at 2025-03-01T07:05:25.878947+00:00
+            # Runnable[2s]: starts at 2025-03-01T07:05:25.879392+00:00
+            # Runnable[3s]: starts at 2025-03-01T07:05:25.879804+00:00
+            # Runnable[2s]: ends at 2025-03-01T07:05:27.881998+00:00
+            # on end callback starts at 2025-03-01T07:05:27.882360+00:00
+            # Runnable[3s]: ends at 2025-03-01T07:05:28.881737+00:00
+            # on end callback starts at 2025-03-01T07:05:28.882428+00:00
+            # on end callback ends at 2025-03-01T07:05:29.883893+00:00
+            # on end callback ends at 2025-03-01T07:05:30.884831+00:00
             ```
         """
         return RunnableBinding(
@@ -1843,7 +1876,7 @@ class Runnable(ABC, Generic[Input, Output]):
                 `exp_base`, and `jitter` (all `float` values).
 
         Returns:
-            A new Runnable that retries the original Runnable on exceptions.
+            A new `Runnable` that retries the original `Runnable` on exceptions.
 
         Example:
             ```python
@@ -1927,7 +1960,9 @@ class Runnable(ABC, Generic[Input, Output]):
             exceptions_to_handle: A tuple of exception types to handle.
             exception_key: If `string` is specified then handled exceptions will be
                 passed to fallbacks as part of the input under the specified key.
+
                 If `None`, exceptions will not be passed to fallbacks.
+
                 If used, the base `Runnable` and its fallbacks must accept a
                 dictionary as input.
 
@@ -1963,7 +1998,9 @@ class Runnable(ABC, Generic[Input, Output]):
             exceptions_to_handle: A tuple of exception types to handle.
             exception_key: If `string` is specified then handled exceptions will be
                 passed to fallbacks as part of the input under the specified key.
+
                 If `None`, exceptions will not be passed to fallbacks.
+
                 If used, the base `Runnable` and its fallbacks must accept a
                 dictionary as input.
 
@@ -2429,10 +2466,14 @@ class Runnable(ABC, Generic[Input, Output]):
 
         `as_tool` will instantiate a `BaseTool` with a name, description, and
         `args_schema` from a `Runnable`. Where possible, schemas are inferred
-        from `runnable.get_input_schema`. Alternatively (e.g., if the
-        `Runnable` takes a dict as input and the specific dict keys are not typed),
-        the schema can be specified directly with `args_schema`. You can also
-        pass `arg_types` to just specify the required arguments and their types.
+        from `runnable.get_input_schema`.
+
+        Alternatively (e.g., if the `Runnable` takes a dict as input and the specific
+        `dict` keys are not typed), the schema can be specified directly with
+        `args_schema`.
+
+        You can also pass `arg_types` to just specify the required arguments and their
+        types.
 
         Args:
             args_schema: The schema for the tool.
@@ -2443,82 +2484,82 @@ class Runnable(ABC, Generic[Input, Output]):
         Returns:
             A `BaseTool` instance.
 
-        Typed dict input:
+        !!! example "`TypedDict` input"
 
-        ```python
-        from typing_extensions import TypedDict
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from typing_extensions import TypedDict
+            from langchain_core.runnables import RunnableLambda
 
 
-        class Args(TypedDict):
-            a: int
-            b: list[int]
+            class Args(TypedDict):
+                a: int
+                b: list[int]
 
 
-        def f(x: Args) -> str:
-            return str(x["a"] * max(x["b"]))
+            def f(x: Args) -> str:
+                return str(x["a"] * max(x["b"]))
 
 
-        runnable = RunnableLambda(f)
-        as_tool = runnable.as_tool()
-        as_tool.invoke({"a": 3, "b": [1, 2]})
-        ```
+            runnable = RunnableLambda(f)
+            as_tool = runnable.as_tool()
+            as_tool.invoke({"a": 3, "b": [1, 2]})
+            ```
 
-        `dict` input, specifying schema via `args_schema`:
+        !!! example "`dict` input, specifying schema via `args_schema`"
 
-        ```python
-        from typing import Any
-        from pydantic import BaseModel, Field
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from typing import Any
+            from pydantic import BaseModel, Field
+            from langchain_core.runnables import RunnableLambda
 
-        def f(x: dict[str, Any]) -> str:
-            return str(x["a"] * max(x["b"]))
+            def f(x: dict[str, Any]) -> str:
+                return str(x["a"] * max(x["b"]))
 
-        class FSchema(BaseModel):
-            \"\"\"Apply a function to an integer and list of integers.\"\"\"
+            class FSchema(BaseModel):
+                \"\"\"Apply a function to an integer and list of integers.\"\"\"
 
-            a: int = Field(..., description="Integer")
-            b: list[int] = Field(..., description="List of ints")
+                a: int = Field(..., description="Integer")
+                b: list[int] = Field(..., description="List of ints")
 
-        runnable = RunnableLambda(f)
-        as_tool = runnable.as_tool(FSchema)
-        as_tool.invoke({"a": 3, "b": [1, 2]})
-        ```
+            runnable = RunnableLambda(f)
+            as_tool = runnable.as_tool(FSchema)
+            as_tool.invoke({"a": 3, "b": [1, 2]})
+            ```
 
-        `dict` input, specifying schema via `arg_types`:
+        !!! example "`dict` input, specifying schema via `arg_types`"
 
-        ```python
-        from typing import Any
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from typing import Any
+            from langchain_core.runnables import RunnableLambda
 
 
-        def f(x: dict[str, Any]) -> str:
-            return str(x["a"] * max(x["b"]))
+            def f(x: dict[str, Any]) -> str:
+                return str(x["a"] * max(x["b"]))
 
 
-        runnable = RunnableLambda(f)
-        as_tool = runnable.as_tool(arg_types={"a": int, "b": list[int]})
-        as_tool.invoke({"a": 3, "b": [1, 2]})
-        ```
+            runnable = RunnableLambda(f)
+            as_tool = runnable.as_tool(arg_types={"a": int, "b": list[int]})
+            as_tool.invoke({"a": 3, "b": [1, 2]})
+            ```
 
-        String input:
+        !!! example "`str` input"
 
-        ```python
-        from langchain_core.runnables import RunnableLambda
+            ```python
+            from langchain_core.runnables import RunnableLambda
 
 
-        def f(x: str) -> str:
-            return x + "a"
+            def f(x: str) -> str:
+                return x + "a"
 
 
-        def g(x: str) -> str:
-            return x + "z"
+            def g(x: str) -> str:
+                return x + "z"
 
 
-        runnable = RunnableLambda(f) | g
-        as_tool = runnable.as_tool()
-        as_tool.invoke("b")
-        ```
+            runnable = RunnableLambda(f) | g
+            as_tool = runnable.as_tool()
+            as_tool.invoke("b")
+            ```
         """
         # Avoid circular import
         from langchain_core.tools import convert_runnable_to_tool  # noqa: PLC0415
@@ -2570,29 +2611,33 @@ class RunnableSerializable(Serializable, Runnable[Input, Output]):
         Returns:
             A new `Runnable` with the fields configured.
 
-        ```python
-        from langchain_core.runnables import ConfigurableField
-        from langchain_openai import ChatOpenAI
+        !!! example
 
-        model = ChatOpenAI(max_tokens=20).configurable_fields(
-            max_tokens=ConfigurableField(
-                id="output_token_number",
-                name="Max tokens in the output",
-                description="The maximum number of tokens in the output",
+            ```python
+            from langchain_core.runnables import ConfigurableField
+            from langchain_openai import ChatOpenAI
+
+            model = ChatOpenAI(max_tokens=20).configurable_fields(
+                max_tokens=ConfigurableField(
+                    id="output_token_number",
+                    name="Max tokens in the output",
+                    description="The maximum number of tokens in the output",
+                )
             )
-        )
 
-        # max_tokens = 20
-        print("max_tokens_20: ", model.invoke("tell me something about chess").content)
+            # max_tokens = 20
+            print(
+                "max_tokens_20: ", model.invoke("tell me something about chess").content
+            )
 
-        # max_tokens = 200
-        print(
-            "max_tokens_200: ",
-            model.with_config(configurable={"output_token_number": 200})
-            .invoke("tell me something about chess")
-            .content,
-        )
-        ```
+            # max_tokens = 200
+            print(
+                "max_tokens_200: ",
+                model.with_config(configurable={"output_token_number": 200})
+                .invoke("tell me something about chess")
+                .content,
+            )
+            ```
         """
         # Import locally to prevent circular import
         from langchain_core.runnables.configurable import (  # noqa: PLC0415
@@ -2631,29 +2676,31 @@ class RunnableSerializable(Serializable, Runnable[Input, Output]):
         Returns:
             A new `Runnable` with the alternatives configured.
 
-        ```python
-        from langchain_anthropic import ChatAnthropic
-        from langchain_core.runnables.utils import ConfigurableField
-        from langchain_openai import ChatOpenAI
+        !!! example
 
-        model = ChatAnthropic(
-            model_name="claude-sonnet-4-5-20250929"
-        ).configurable_alternatives(
-            ConfigurableField(id="llm"),
-            default_key="anthropic",
-            openai=ChatOpenAI(),
-        )
+            ```python
+            from langchain_anthropic import ChatAnthropic
+            from langchain_core.runnables.utils import ConfigurableField
+            from langchain_openai import ChatOpenAI
 
-        # uses the default model ChatAnthropic
-        print(model.invoke("which organization created you?").content)
+            model = ChatAnthropic(
+                model_name="claude-sonnet-4-5-20250929"
+            ).configurable_alternatives(
+                ConfigurableField(id="llm"),
+                default_key="anthropic",
+                openai=ChatOpenAI(),
+            )
 
-        # uses ChatOpenAI
-        print(
-            model.with_config(configurable={"llm": "openai"})
-            .invoke("which organization created you?")
-            .content
-        )
-        ```
+            # uses the default model ChatAnthropic
+            print(model.invoke("which organization created you?").content)
+
+            # uses ChatOpenAI
+            print(
+                model.with_config(configurable={"llm": "openai"})
+                .invoke("which organization created you?")
+                .content
+            )
+            ```
         """
         # Import locally to prevent circular import
         from langchain_core.runnables.configurable import (  # noqa: PLC0415
@@ -2750,6 +2797,9 @@ def _seq_output_schema(
     return last.get_output_schema(config)
 
 
+_RUNNABLE_SEQUENCE_MIN_STEPS = 2
+
+
 class RunnableSequence(RunnableSerializable[Input, Output]):
     """Sequence of `Runnable` objects, where the output of one is the input of the next.
 
@@ -2859,7 +2909,7 @@ class RunnableSequence(RunnableSerializable[Input, Output]):
             name: The name of the `Runnable`.
             first: The first `Runnable` in the sequence.
             middle: The middle `Runnable` objects in the sequence.
-            last: The last Runnable in the sequence.
+            last: The last `Runnable` in the sequence.
 
         Raises:
             ValueError: If the sequence has less than 2 steps.
@@ -2872,8 +2922,11 @@ class RunnableSequence(RunnableSerializable[Input, Output]):
                 steps_flat.extend(step.steps)
             else:
                 steps_flat.append(coerce_to_runnable(step))
-        if len(steps_flat) < 2:
-            msg = f"RunnableSequence must have at least 2 steps, got {len(steps_flat)}"
+        if len(steps_flat) < _RUNNABLE_SEQUENCE_MIN_STEPS:
+            msg = (
+                f"RunnableSequence must have at least {_RUNNABLE_SEQUENCE_MIN_STEPS} "
+                f"steps, got {len(steps_flat)}"
+            )
             raise ValueError(msg)
         super().__init__(
             first=steps_flat[0],
@@ -2904,7 +2957,7 @@ class RunnableSequence(RunnableSerializable[Input, Output]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     model_config = ConfigDict(
@@ -3610,7 +3663,7 @@ class RunnableParallel(RunnableSerializable[Input, dict[str, Any]]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -3668,6 +3721,12 @@ class RunnableParallel(RunnableSerializable[Input, dict[str, Any]]):
             == "object"
             for s in self.steps__.values()
         ):
+            for step in self.steps__.values():
+                fields = step.get_input_schema(config).model_fields
+                root_field = fields.get("root")
+                if root_field is not None and root_field.annotation != Any:
+                    return super().get_input_schema(config)
+
             # This is correct, but pydantic typings/mypy don't think so.
             return create_model_v2(
                 self.get_name("Input"),
@@ -4477,7 +4536,7 @@ class RunnableLambda(Runnable[Input, Output]):
             # on itemgetter objects, so we have to parse the repr
             items = str(func).replace("operator.itemgetter(", "")[:-1].split(", ")
             if all(
-                item[0] == "'" and item[-1] == "'" and len(item) > 2 for item in items
+                item[0] == "'" and item[-1] == "'" and item != "''" for item in items
             ):
                 fields = {item[1:-1]: (Any, ...) for item in items}
                 # It's a dict, lol
@@ -5139,7 +5198,7 @@ class RunnableEachBase(RunnableSerializable[list[Input], list[Output]]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -5322,7 +5381,7 @@ class RunnableEach(RunnableEachBase[Input, Output]):
 
 
 class RunnableBindingBase(RunnableSerializable[Input, Output]):  # type: ignore[no-redef]
-    """`Runnable` that delegates calls to another `Runnable` with a set of kwargs.
+    """`Runnable` that delegates calls to another `Runnable` with a set of `**kwargs`.
 
     Use only if creating a new `RunnableBinding` subclass with different `__init__`
     args.
@@ -5462,7 +5521,7 @@ class RunnableBindingBase(RunnableSerializable[Input, Output]):  # type: ignore[
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -5752,7 +5811,7 @@ class RunnableBinding(RunnableBindingBase[Input, Output]):  # type: ignore[no-re
         ```python
         # Create a Runnable binding that invokes the chat model with the
         # additional kwarg `stop=['-']` when running it.
-        from langchain_community.chat_models import ChatOpenAI
+        from langchain_openai import ChatOpenAI
 
         model = ChatOpenAI()
         model.invoke('Say "Parrot-MAGIC"', stop=["-"])  # Should return `Parrot`

libs/core/langchain_core/runnables/branch.py

@@ -36,11 +36,13 @@ from langchain_core.runnables.utils import (
     get_unique_config_specs,
 )
 
+_MIN_BRANCHES = 2
+
 
 class RunnableBranch(RunnableSerializable[Input, Output]):
-    """Runnable that selects which branch to run based on a condition.
+    """`Runnable` that selects which branch to run based on a condition.
 
-    The Runnable is initialized with a list of `(condition, Runnable)` pairs and
+    The `Runnable` is initialized with a list of `(condition, Runnable)` pairs and
     a default branch.
 
     When operating on an input, the first condition that evaluates to True is
@@ -86,12 +88,12 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
                 Defaults a `Runnable` to run if no condition is met.
 
         Raises:
-            ValueError: If the number of branches is less than 2.
+            ValueError: If the number of branches is less than `2`.
             TypeError: If the default branch is not `Runnable`, `Callable` or `Mapping`.
-            TypeError: If a branch is not a tuple or list.
-            ValueError: If a branch is not of length 2.
+            TypeError: If a branch is not a `tuple` or `list`.
+            ValueError: If a branch is not of length `2`.
         """
-        if len(branches) < 2:
+        if len(branches) < _MIN_BRANCHES:
             msg = "RunnableBranch requires at least two branches"
             raise ValueError(msg)
 
@@ -118,7 +120,7 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
                 )
                 raise TypeError(msg)
 
-            if len(branch) != 2:
+            if len(branch) != _MIN_BRANCHES:
                 msg = (
                     f"RunnableBranch branches must be "
                     f"tuples or lists of length 2, not {len(branch)}"
@@ -140,7 +142,7 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
 
     @classmethod
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -187,12 +189,12 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
     def invoke(
         self, input: Input, config: RunnableConfig | None = None, **kwargs: Any
     ) -> Output:
-        """First evaluates the condition, then delegate to true or false branch.
+        """First evaluates the condition, then delegate to `True` or `False` branch.
 
         Args:
-            input: The input to the Runnable.
-            config: The configuration for the Runnable.
-            **kwargs: Additional keyword arguments to pass to the Runnable.
+            input: The input to the `Runnable`.
+            config: The configuration for the `Runnable`.
+            **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
         Returns:
             The output of the branch that was run.
@@ -297,12 +299,12 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
         config: RunnableConfig | None = None,
         **kwargs: Any | None,
     ) -> Iterator[Output]:
-        """First evaluates the condition, then delegate to true or false branch.
+        """First evaluates the condition, then delegate to `True` or `False` branch.
 
         Args:
-            input: The input to the Runnable.
-            config: The configuration for the Runnable.
-            **kwargs: Additional keyword arguments to pass to the Runnable.
+            input: The input to the `Runnable`.
+            config: The configuration for the `Runnable`.
+            **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
         Yields:
             The output of the branch that was run.
@@ -381,12 +383,12 @@ class RunnableBranch(RunnableSerializable[Input, Output]):
         config: RunnableConfig | None = None,
         **kwargs: Any | None,
     ) -> AsyncIterator[Output]:
-        """First evaluates the condition, then delegate to true or false branch.
+        """First evaluates the condition, then delegate to `True` or `False` branch.
 
         Args:
-            input: The input to the Runnable.
-            config: The configuration for the Runnable.
-            **kwargs: Additional keyword arguments to pass to the Runnable.
+            input: The input to the `Runnable`.
+            config: The configuration for the `Runnable`.
+            **kwargs: Additional keyword arguments to pass to the `Runnable`.
 
         Yields:
             The output of the branch that was run.

libs/core/langchain_core/runnables/config.py

@@ -47,54 +47,59 @@ class EmptyDict(TypedDict, total=False):
 
 
 class RunnableConfig(TypedDict, total=False):
-    """Configuration for a Runnable."""
+    """Configuration for a `Runnable`.
+
+    See the [reference docs](https://reference.langchain.com/python/langchain_core/runnables/#langchain_core.runnables.RunnableConfig)
+    for more details.
+    """
 
     tags: list[str]
-    """
-    Tags for this call and any sub-calls (eg. a Chain calling an LLM).
+    """Tags for this call and any sub-calls (e.g. a Chain calling an LLM).
+
     You can use these to filter calls.
     """
 
     metadata: dict[str, Any]
-    """
-    Metadata for this call and any sub-calls (eg. a Chain calling an LLM).
+    """Metadata for this call and any sub-calls (e.g. a Chain calling an LLM).
+
     Keys should be strings, values should be JSON-serializable.
     """
 
     callbacks: Callbacks
-    """
-    Callbacks for this call and any sub-calls (eg. a Chain calling an LLM).
+    """Callbacks for this call and any sub-calls (e.g. a Chain calling an LLM).
+
     Tags are passed to all callbacks, metadata is passed to handle*Start callbacks.
     """
 
     run_name: str
-    """
-    Name for the tracer run for this call. Defaults to the name of the class.
-    """
+    """Name for the tracer run for this call.
+
+    Defaults to the name of the class."""
 
     max_concurrency: int | None
-    """
-    Maximum number of parallel calls to make. If not provided, defaults to
-    `ThreadPoolExecutor`'s default.
+    """Maximum number of parallel calls to make.
+
+    If not provided, defaults to `ThreadPoolExecutor`'s default.
     """
 
     recursion_limit: int
-    """
-    Maximum number of times a call can recurse. If not provided, defaults to `25`.
+    """Maximum number of times a call can recurse.
+
+    If not provided, defaults to `25`.
     """
 
     configurable: dict[str, Any]
-    """
-    Runtime values for attributes previously made configurable on this `Runnable`,
+    """Runtime values for attributes previously made configurable on this `Runnable`,
     or sub-Runnables, through `configurable_fields` or `configurable_alternatives`.
+
     Check `output_schema` for a description of the attributes that have been made
     configurable.
     """
 
     run_id: uuid.UUID | None
-    """
-    Unique identifier for the tracer run for this call. If not provided, a new UUID
-    will be generated.
+    """Unique identifier for the tracer run for this call.
+
+    If not provided, a new UUID will be generated.
     """
 
 

libs/core/langchain_core/runnables/configurable.py

@@ -1,4 +1,4 @@
-"""Runnables that can be dynamically configured."""
+"""`Runnable` objects that can be dynamically configured."""
 
 from __future__ import annotations
 
@@ -47,14 +47,14 @@ if TYPE_CHECKING:
 
 
 class DynamicRunnable(RunnableSerializable[Input, Output]):
-    """Serializable Runnable that can be dynamically configured.
+    """Serializable `Runnable` that can be dynamically configured.
 
-    A DynamicRunnable should be initiated using the `configurable_fields` or
-    `configurable_alternatives` method of a Runnable.
+    A `DynamicRunnable` should be initiated using the `configurable_fields` or
+    `configurable_alternatives` method of a `Runnable`.
     """
 
     default: RunnableSerializable[Input, Output]
-    """The default Runnable to use."""
+    """The default `Runnable` to use."""
 
     config: RunnableConfig | None = None
     """The configuration to use."""
@@ -66,7 +66,7 @@ class DynamicRunnable(RunnableSerializable[Input, Output]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -120,13 +120,13 @@ class DynamicRunnable(RunnableSerializable[Input, Output]):
     def prepare(
         self, config: RunnableConfig | None = None
     ) -> tuple[Runnable[Input, Output], RunnableConfig]:
-        """Prepare the Runnable for invocation.
+        """Prepare the `Runnable` for invocation.
 
         Args:
             config: The configuration to use.
 
         Returns:
-            The prepared Runnable and configuration.
+            The prepared `Runnable` and configuration.
         """
         runnable: Runnable[Input, Output] = self
         while isinstance(runnable, DynamicRunnable):
@@ -316,12 +316,12 @@ class DynamicRunnable(RunnableSerializable[Input, Output]):
 
 
 class RunnableConfigurableFields(DynamicRunnable[Input, Output]):
-    """Runnable that can be dynamically configured.
+    """`Runnable` that can be dynamically configured.
 
-    A RunnableConfigurableFields should be initiated using the
-    `configurable_fields` method of a Runnable.
+    A `RunnableConfigurableFields` should be initiated using the
+    `configurable_fields` method of a `Runnable`.
 
-    Here is an example of using a RunnableConfigurableFields with LLMs:
+    Here is an example of using a `RunnableConfigurableFields` with LLMs:
 
         ```python
         from langchain_core.prompts import PromptTemplate
@@ -348,7 +348,7 @@ class RunnableConfigurableFields(DynamicRunnable[Input, Output]):
         chain.invoke({"x": 0}, config={"configurable": {"temperature": 0.9}})
         ```
 
-    Here is an example of using a RunnableConfigurableFields with HubRunnables:
+    Here is an example of using a `RunnableConfigurableFields` with `HubRunnables`:
 
         ```python
         from langchain_core.prompts import PromptTemplate
@@ -380,7 +380,7 @@ class RunnableConfigurableFields(DynamicRunnable[Input, Output]):
 
     @property
     def config_specs(self) -> list[ConfigurableFieldSpec]:
-        """Get the configuration specs for the RunnableConfigurableFields.
+        """Get the configuration specs for the `RunnableConfigurableFields`.
 
         Returns:
             The configuration specs.
@@ -473,10 +473,10 @@ _enums_for_spec_lock = threading.Lock()
 
 
 class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
-    """Runnable that can be dynamically configured.
+    """`Runnable` that can be dynamically configured.
 
     A `RunnableConfigurableAlternatives` should be initiated using the
-    `configurable_alternatives` method of a Runnable or can be
+    `configurable_alternatives` method of a `Runnable` or can be
     initiated directly as well.
 
     Here is an example of using a `RunnableConfigurableAlternatives` that uses
@@ -531,7 +531,7 @@ class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
     """
 
     which: ConfigurableField
-    """The ConfigurableField to use to choose between alternatives."""
+    """The `ConfigurableField` to use to choose between alternatives."""
 
     alternatives: dict[
         str,
@@ -544,8 +544,9 @@ class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
 
     prefix_keys: bool
     """Whether to prefix configurable fields of each alternative with a namespace
-    of the form <which.id>==<alternative_key>, eg. a key named "temperature" used by
-    the alternative named "gpt3" becomes "model==gpt3/temperature"."""
+    of the form <which.id>==<alternative_key>, e.g. a key named "temperature" used by
+    the alternative named "gpt3" becomes "model==gpt3/temperature".
+    """
 
     @property
     @override
@@ -638,24 +639,24 @@ class RunnableConfigurableAlternatives(DynamicRunnable[Input, Output]):
 
 
 def _strremoveprefix(s: str, prefix: str) -> str:
-    """str.removeprefix() is only available in Python 3.9+."""
+    """`str.removeprefix()` is only available in Python 3.9+."""
     return s.replace(prefix, "", 1) if s.startswith(prefix) else s
 
 
 def prefix_config_spec(
     spec: ConfigurableFieldSpec, prefix: str
 ) -> ConfigurableFieldSpec:
-    """Prefix the id of a ConfigurableFieldSpec.
+    """Prefix the id of a `ConfigurableFieldSpec`.
 
-    This is useful when a RunnableConfigurableAlternatives is used as a
-    ConfigurableField of another RunnableConfigurableAlternatives.
+    This is useful when a `RunnableConfigurableAlternatives` is used as a
+    `ConfigurableField` of another `RunnableConfigurableAlternatives`.
 
     Args:
-        spec: The ConfigurableFieldSpec to prefix.
+        spec: The `ConfigurableFieldSpec` to prefix.
         prefix: The prefix to add.
 
     Returns:
-        The prefixed ConfigurableFieldSpec.
+        The prefixed `ConfigurableFieldSpec`.
     """
     return (
         ConfigurableFieldSpec(
@@ -677,15 +678,15 @@ def make_options_spec(
 ) -> ConfigurableFieldSpec:
     """Make options spec.
 
-    Make a ConfigurableFieldSpec for a ConfigurableFieldSingleOption or
-    ConfigurableFieldMultiOption.
+    Make a `ConfigurableFieldSpec` for a `ConfigurableFieldSingleOption` or
+    `ConfigurableFieldMultiOption`.
 
     Args:
-        spec: The ConfigurableFieldSingleOption or ConfigurableFieldMultiOption.
+        spec: The `ConfigurableFieldSingleOption` or `ConfigurableFieldMultiOption`.
         description: The description to use if the spec does not have one.
 
     Returns:
-        The ConfigurableFieldSpec.
+        The `ConfigurableFieldSpec`.
     """
     with _enums_for_spec_lock:
         if enum := _enums_for_spec.get(spec):

libs/core/langchain_core/runnables/fallbacks.py

@@ -35,20 +35,20 @@ if TYPE_CHECKING:
 
 
 class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
-    """Runnable that can fallback to other Runnables if it fails.
+    """`Runnable` that can fallback to other `Runnable`s if it fails.
 
     External APIs (e.g., APIs for a language model) may at times experience
     degraded performance or even downtime.
 
-    In these cases, it can be useful to have a fallback Runnable that can be
-    used in place of the original Runnable (e.g., fallback to another LLM provider).
+    In these cases, it can be useful to have a fallback `Runnable` that can be
+    used in place of the original `Runnable` (e.g., fallback to another LLM provider).
 
-    Fallbacks can be defined at the level of a single Runnable, or at the level
-    of a chain of Runnables. Fallbacks are tried in order until one succeeds or
+    Fallbacks can be defined at the level of a single `Runnable`, or at the level
+    of a chain of `Runnable`s. Fallbacks are tried in order until one succeeds or
     all fail.
 
     While you can instantiate a `RunnableWithFallbacks` directly, it is usually
-    more convenient to use the `with_fallbacks` method on a Runnable.
+    more convenient to use the `with_fallbacks` method on a `Runnable`.
 
     Example:
         ```python
@@ -87,7 +87,7 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
     """
 
     runnable: Runnable[Input, Output]
-    """The Runnable to run first."""
+    """The `Runnable` to run first."""
     fallbacks: Sequence[Runnable[Input, Output]]
     """A sequence of fallbacks to try."""
     exceptions_to_handle: tuple[type[BaseException], ...] = (Exception,)
@@ -97,9 +97,12 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
     """
     exception_key: str | None = None
     """If `string` is specified then handled exceptions will be passed to fallbacks as
-        part of the input under the specified key. If `None`, exceptions
-        will not be passed to fallbacks. If used, the base Runnable and its fallbacks
-        must accept a dictionary as input."""
+    part of the input under the specified key.
+
+    If `None`, exceptions will not be passed to fallbacks.
+
+    If used, the base `Runnable` and its fallbacks must accept a dictionary as input.
+    """
 
     model_config = ConfigDict(
         arbitrary_types_allowed=True,
@@ -137,7 +140,7 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -152,10 +155,10 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
 
     @property
     def runnables(self) -> Iterator[Runnable[Input, Output]]:
-        """Iterator over the Runnable and its fallbacks.
+        """Iterator over the `Runnable` and its fallbacks.
 
         Yields:
-            The Runnable then its fallbacks.
+            The `Runnable` then its fallbacks.
         """
         yield self.runnable
         yield from self.fallbacks
@@ -589,14 +592,14 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
         await run_manager.on_chain_end(output)
 
     def __getattr__(self, name: str) -> Any:
-        """Get an attribute from the wrapped Runnable and its fallbacks.
+        """Get an attribute from the wrapped `Runnable` and its fallbacks.
 
         Returns:
-            If the attribute is anything other than a method that outputs a Runnable,
-            returns getattr(self.runnable, name). If the attribute is a method that
-            does return a new Runnable (e.g. model.bind_tools([...]) outputs a new
-            RunnableBinding) then self.runnable and each of the runnables in
-            self.fallbacks is replaced with getattr(x, name).
+            If the attribute is anything other than a method that outputs a `Runnable`,
+            returns `getattr(self.runnable, name)`. If the attribute is a method that
+            does return a new `Runnable` (e.g. `model.bind_tools([...])` outputs a new
+            `RunnableBinding`) then `self.runnable` and each of the runnables in
+            `self.fallbacks` is replaced with `getattr(x, name)`.
 
         Example:
             ```python
@@ -618,7 +621,6 @@ class RunnableWithFallbacks(RunnableSerializable[Input, Output]):
                 runnable=RunnableBinding(bound=ChatOpenAI(...), kwargs={"tools": [...]}),
                 fallbacks=[RunnableBinding(bound=ChatAnthropic(...), kwargs={"tools": [...]})],
             )
-
             ```
         """  # noqa: E501
         attr = getattr(self.runnable, name)

libs/core/langchain_core/runnables/graph.py

@@ -4,7 +4,6 @@ from __future__ import annotations
 
 import inspect
 from collections import defaultdict
-from collections.abc import Callable
 from dataclasses import dataclass, field
 from enum import Enum
 from typing import (
@@ -22,7 +21,7 @@ from langchain_core.runnables.base import Runnable, RunnableSerializable
 from langchain_core.utils.pydantic import _IgnoreUnserializable, is_basemodel_subclass
 
 if TYPE_CHECKING:
-    from collections.abc import Sequence
+    from collections.abc import Callable, Sequence
 
     from pydantic import BaseModel
 
@@ -642,6 +641,7 @@ class Graph:
         retry_delay: float = 1.0,
         frontmatter_config: dict[str, Any] | None = None,
         base_url: str | None = None,
+        proxies: dict[str, str] | None = None,
     ) -> bytes:
         """Draw the graph as a PNG image using Mermaid.
 
@@ -674,11 +674,10 @@ class Graph:
                 }
                 ```
             base_url: The base URL of the Mermaid server for rendering via API.
-
+            proxies: HTTP/HTTPS proxies for requests (e.g. `{"http": "http://127.0.0.1:7890"}`).
 
         Returns:
             The PNG image as bytes.
-
         """
         # Import locally to prevent circular import
         from langchain_core.runnables.graph_mermaid import (  # noqa: PLC0415
@@ -699,6 +698,7 @@ class Graph:
             padding=padding,
             max_retries=max_retries,
             retry_delay=retry_delay,
+            proxies=proxies,
             base_url=base_url,
         )
 

libs/core/langchain_core/runnables/graph_ascii.py

@@ -7,7 +7,6 @@ from __future__ import annotations
 
 import math
 import os
-from collections.abc import Mapping, Sequence
 from typing import TYPE_CHECKING, Any
 
 try:
@@ -20,6 +19,8 @@ except ImportError:
     _HAS_GRANDALF = False
 
 if TYPE_CHECKING:
+    from collections.abc import Mapping, Sequence
+
     from langchain_core.runnables.graph import Edge as LangEdge
 
 

libs/core/langchain_core/runnables/graph_mermaid.py

@@ -281,6 +281,7 @@ def draw_mermaid_png(
     max_retries: int = 1,
     retry_delay: float = 1.0,
     base_url: str | None = None,
+    proxies: dict[str, str] | None = None,
 ) -> bytes:
     """Draws a Mermaid graph as PNG using provided syntax.
 
@@ -293,6 +294,7 @@ def draw_mermaid_png(
         max_retries: Maximum number of retries (MermaidDrawMethod.API).
         retry_delay: Delay between retries (MermaidDrawMethod.API).
         base_url: Base URL for the Mermaid.ink API.
+        proxies: HTTP/HTTPS proxies for requests (e.g. `{"http": "http://127.0.0.1:7890"}`).
 
     Returns:
         PNG image bytes.
@@ -314,6 +316,7 @@ def draw_mermaid_png(
             max_retries=max_retries,
             retry_delay=retry_delay,
             base_url=base_url,
+            proxies=proxies,
         )
     else:
         supported_methods = ", ".join([m.value for m in MermaidDrawMethod])
@@ -405,6 +408,7 @@ def _render_mermaid_using_api(
     file_type: Literal["jpeg", "png", "webp"] | None = "png",
     max_retries: int = 1,
     retry_delay: float = 1.0,
+    proxies: dict[str, str] | None = None,
     base_url: str | None = None,
 ) -> bytes:
     """Renders Mermaid graph using the Mermaid.INK API."""
@@ -445,7 +449,7 @@ def _render_mermaid_using_api(
 
     for attempt in range(max_retries + 1):
         try:
-            response = requests.get(image_url, timeout=10)
+            response = requests.get(image_url, timeout=10, proxies=proxies)
             if response.status_code == requests.codes.ok:
                 img_bytes = response.content
                 if output_file_path is not None:
@@ -454,7 +458,10 @@ def _render_mermaid_using_api(
                 return img_bytes
 
             # If we get a server error (5xx), retry
-            if 500 <= response.status_code < 600 and attempt < max_retries:
+            if (
+                requests.codes.internal_server_error <= response.status_code
+                and attempt < max_retries
+            ):
                 # Exponential backoff with jitter
                 sleep_time = retry_delay * (2**attempt) * (0.5 + 0.5 * random.random())  # noqa: S311 not used for crypto
                 time.sleep(sleep_time)

libs/core/langchain_core/runnables/graph_png.py

@@ -1,5 +1,6 @@
 """Helper class to draw a state graph into a PNG file."""
 
+from itertools import groupby
 from typing import Any
 
 from langchain_core.runnables.graph import Graph, LabelsDict
@@ -141,6 +142,7 @@ class PngDrawer:
         # Add nodes, conditional edges, and edges to the graph
         self.add_nodes(viz, graph)
         self.add_edges(viz, graph)
+        self.add_subgraph(viz, [node.split(":") for node in graph.nodes])
 
         # Update entrypoint and END styles
         self.update_styles(viz, graph)
@@ -161,6 +163,32 @@ class PngDrawer:
         for node in graph.nodes:
             self.add_node(viz, node)
 
+    def add_subgraph(
+        self,
+        viz: Any,
+        nodes: list[list[str]],
+        parent_prefix: list[str] | None = None,
+    ) -> None:
+        """Add subgraphs to the graph.
+
+        Args:
+            viz: The graphviz object.
+            nodes: The nodes to add.
+            parent_prefix: The prefix of the parent subgraph.
+        """
+        for prefix, grouped in groupby(
+            [node[:] for node in sorted(nodes)],
+            key=lambda x: x.pop(0),
+        ):
+            current_prefix = (parent_prefix or []) + [prefix]
+            grouped_nodes = list(grouped)
+            if len(grouped_nodes) > 1:
+                subgraph = viz.add_subgraph(
+                    [":".join(current_prefix + node) for node in grouped_nodes],
+                    name="cluster_" + ":".join(current_prefix),
+                )
+                self.add_subgraph(subgraph, grouped_nodes, current_prefix)
+
     def add_edges(self, viz: Any, graph: Graph) -> None:
         """Add edges to the graph.
 

libs/core/langchain_core/runnables/history.py

@@ -36,23 +36,23 @@ GetSessionHistoryCallable = Callable[..., BaseChatMessageHistory]
 
 
 class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
-    """Runnable that manages chat message history for another Runnable.
+    """`Runnable` that manages chat message history for another `Runnable`.
 
     A chat message history is a sequence of messages that represent a conversation.
 
-    RunnableWithMessageHistory wraps another Runnable and manages the chat message
+    `RunnableWithMessageHistory` wraps another `Runnable` and manages the chat message
     history for it; it is responsible for reading and updating the chat message
     history.
 
-    The formats supported for the inputs and outputs of the wrapped Runnable
+    The formats supported for the inputs and outputs of the wrapped `Runnable`
     are described below.
 
-    RunnableWithMessageHistory must always be called with a config that contains
+    `RunnableWithMessageHistory` must always be called with a config that contains
     the appropriate parameters for the chat message history factory.
 
-    By default, the Runnable is expected to take a single configuration parameter
+    By default, the `Runnable` is expected to take a single configuration parameter
     called `session_id` which is a string. This parameter is used to create a new
-    or look up an existing chat message history that matches the given session_id.
+    or look up an existing chat message history that matches the given `session_id`.
 
     In this case, the invocation would look like this:
 
@@ -117,12 +117,12 @@ class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
 
         ```
 
-    Example where the wrapped Runnable takes a dictionary input:
+    Example where the wrapped `Runnable` takes a dictionary input:
 
         ```python
         from typing import Optional
 
-        from langchain_community.chat_models import ChatAnthropic
+        from langchain_anthropic import ChatAnthropic
         from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
         from langchain_core.runnables.history import RunnableWithMessageHistory
 
@@ -166,7 +166,7 @@ class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
         print(store)  # noqa: T201
         ```
 
-    Example where the session factory takes two keys, user_id and conversation id):
+    Example where the session factory takes two keys (`user_id` and `conversation_id`):
 
         ```python
         store = {}
@@ -223,21 +223,28 @@ class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
     """
 
     get_session_history: GetSessionHistoryCallable
-    """Function that returns a new BaseChatMessageHistory.
+    """Function that returns a new `BaseChatMessageHistory`.
+
     This function should either take a single positional argument `session_id` of type
-    string and return a corresponding chat message history instance"""
+    string and return a corresponding chat message history instance
+    """
     input_messages_key: str | None = None
-    """Must be specified if the base runnable accepts a dict as input.
-    The key in the input dict that contains the messages."""
+    """Must be specified if the base `Runnable` accepts a `dict` as input.
+    The key in the input `dict` that contains the messages.
+    """
     output_messages_key: str | None = None
-    """Must be specified if the base Runnable returns a dict as output.
-    The key in the output dict that contains the messages."""
+    """Must be specified if the base `Runnable` returns a `dict` as output.
+    The key in the output `dict` that contains the messages.
+    """
     history_messages_key: str | None = None
-    """Must be specified if the base runnable accepts a dict as input and expects a
-    separate key for historical messages."""
+    """Must be specified if the base `Runnable` accepts a `dict` as input and expects a
+    separate key for historical messages.
+    """
     history_factory_config: Sequence[ConfigurableFieldSpec]
     """Configure fields that should be passed to the chat history factory.
-    See `ConfigurableFieldSpec` for more details."""
+
+    See `ConfigurableFieldSpec` for more details.
+    """
 
     def __init__(
         self,
@@ -254,15 +261,16 @@ class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
         history_factory_config: Sequence[ConfigurableFieldSpec] | None = None,
         **kwargs: Any,
     ) -> None:
-        """Initialize RunnableWithMessageHistory.
+        """Initialize `RunnableWithMessageHistory`.
 
         Args:
-            runnable: The base Runnable to be wrapped.
+            runnable: The base `Runnable` to be wrapped.
+
                 Must take as input one of:
 
                 1. A list of `BaseMessage`
-                2. A dict with one key for all messages
-                3. A dict with one key for the current input string/message(s) and
+                2. A `dict` with one key for all messages
+                3. A `dict` with one key for the current input string/message(s) and
                     a separate key for historical messages. If the input key points
                     to a string, it will be treated as a `HumanMessage` in history.
 
@@ -270,13 +278,15 @@ class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
 
                 1. A string which can be treated as an `AIMessage`
                 2. A `BaseMessage` or sequence of `BaseMessage`
-                3. A dict with a key for a `BaseMessage` or sequence of
+                3. A `dict` with a key for a `BaseMessage` or sequence of
                     `BaseMessage`
 
-            get_session_history: Function that returns a new BaseChatMessageHistory.
+            get_session_history: Function that returns a new `BaseChatMessageHistory`.
+
                 This function should either take a single positional argument
                 `session_id` of type string and return a corresponding
                 chat message history instance.
+
                 ```python
                 def get_session_history(
                     session_id: str, *, user_id: str | None = None
@@ -295,16 +305,17 @@ class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
                 ) -> BaseChatMessageHistory: ...
                 ```
 
-            input_messages_key: Must be specified if the base runnable accepts a dict
+            input_messages_key: Must be specified if the base runnable accepts a `dict`
                 as input.
-            output_messages_key: Must be specified if the base runnable returns a dict
+            output_messages_key: Must be specified if the base runnable returns a `dict`
                 as output.
-            history_messages_key: Must be specified if the base runnable accepts a dict
-                as input and expects a separate key for historical messages.
+            history_messages_key: Must be specified if the base runnable accepts a
+                `dict` as input and expects a separate key for historical messages.
             history_factory_config: Configure fields that should be passed to the
                 chat history factory. See `ConfigurableFieldSpec` for more details.
-                Specifying these allows you to pass multiple config keys
-                into the get_session_history factory.
+
+                Specifying these allows you to pass multiple config keys into the
+                `get_session_history` factory.
             **kwargs: Arbitrary additional kwargs to pass to parent class
                 `RunnableBindingBase` init.
 
@@ -364,7 +375,7 @@ class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
     @property
     @override
     def config_specs(self) -> list[ConfigurableFieldSpec]:
-        """Get the configuration specs for the RunnableWithMessageHistory."""
+        """Get the configuration specs for the `RunnableWithMessageHistory`."""
         return get_unique_config_specs(
             super().config_specs + list(self.history_factory_config)
         )
@@ -606,6 +617,6 @@ class RunnableWithMessageHistory(RunnableBindingBase):  # type: ignore[no-redef]
 
 
 def _get_parameter_names(callable_: GetSessionHistoryCallable) -> list[str]:
-    """Get the parameter names of the callable."""
+    """Get the parameter names of the `Callable`."""
     sig = inspect.signature(callable_)
     return list(sig.parameters.keys())

libs/core/langchain_core/runnables/passthrough.py

@@ -51,10 +51,10 @@ def identity(x: Other) -> Other:
     """Identity function.
 
     Args:
-        x: input.
+        x: Input.
 
     Returns:
-        output.
+        Output.
     """
     return x
 
@@ -63,10 +63,10 @@ async def aidentity(x: Other) -> Other:
     """Async identity function.
 
     Args:
-        x: input.
+        x: Input.
 
     Returns:
-        output.
+        Output.
     """
     return x
 
@@ -74,11 +74,11 @@ async def aidentity(x: Other) -> Other:
 class RunnablePassthrough(RunnableSerializable[Other, Other]):
     """Runnable to passthrough inputs unchanged or with additional keys.
 
-    This Runnable behaves almost like the identity function, except that it
+    This `Runnable` behaves almost like the identity function, except that it
     can be configured to add additional keys to the output, if the input is a
     dict.
 
-    The examples below demonstrate this Runnable works using a few simple
+    The examples below demonstrate this `Runnable` works using a few simple
     chains. The chains rely on simple lambdas to make the examples easy to execute
     and experiment with.
 
@@ -164,7 +164,7 @@ class RunnablePassthrough(RunnableSerializable[Other, Other]):
         input_type: type[Other] | None = None,
         **kwargs: Any,
     ) -> None:
-        """Create e RunnablePassthrough.
+        """Create a `RunnablePassthrough`.
 
         Args:
             func: Function to be called with the input.
@@ -180,7 +180,7 @@ class RunnablePassthrough(RunnableSerializable[Other, Other]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -213,11 +213,11 @@ class RunnablePassthrough(RunnableSerializable[Other, Other]):
         """Merge the Dict input with the output produced by the mapping argument.
 
         Args:
-            **kwargs: Runnable, Callable or a Mapping from keys to Runnables
-                or Callables.
+            **kwargs: `Runnable`, `Callable` or a `Mapping` from keys to `Runnable`
+                objects or `Callable`s.
 
         Returns:
-            A Runnable that merges the Dict input with the output produced by the
+            A `Runnable` that merges the `dict` input with the output produced by the
             mapping argument.
         """
         return RunnableAssign(RunnableParallel[dict[str, Any]](kwargs))
@@ -350,7 +350,7 @@ _graph_passthrough: RunnablePassthrough = RunnablePassthrough()
 
 
 class RunnableAssign(RunnableSerializable[dict[str, Any], dict[str, Any]]):
-    """Runnable that assigns key-value pairs to dict[str, Any] inputs.
+    """Runnable that assigns key-value pairs to `dict[str, Any]` inputs.
 
     The `RunnableAssign` class takes input dictionaries and, through a
     `RunnableParallel` instance, applies transformations, then combines
@@ -392,7 +392,7 @@ class RunnableAssign(RunnableSerializable[dict[str, Any], dict[str, Any]]):
     mapper: RunnableParallel
 
     def __init__(self, mapper: RunnableParallel[dict[str, Any]], **kwargs: Any) -> None:
-        """Create a RunnableAssign.
+        """Create a `RunnableAssign`.
 
         Args:
             mapper: A `RunnableParallel` instance that will be used to transform the
@@ -403,7 +403,7 @@ class RunnableAssign(RunnableSerializable[dict[str, Any], dict[str, Any]]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod
@@ -668,13 +668,19 @@ class RunnableAssign(RunnableSerializable[dict[str, Any], dict[str, Any]]):
             yield chunk
 
 
-class RunnablePick(RunnableSerializable[dict[str, Any], dict[str, Any]]):
-    """Runnable that picks keys from dict[str, Any] inputs.
+class RunnablePick(RunnableSerializable[dict[str, Any], Any]):
+    """`Runnable` that picks keys from `dict[str, Any]` inputs.
 
-    RunnablePick class represents a Runnable that selectively picks keys from a
+    `RunnablePick` class represents a `Runnable` that selectively picks keys from a
     dictionary input. It allows you to specify one or more keys to extract
-    from the input dictionary. It returns a new dictionary containing only
-    the selected keys.
+    from the input dictionary.
+
+    !!! note "Return Type Behavior"
+        The return type depends on the `keys` parameter:
+
+        - When `keys` is a `str`: Returns the single value associated with that key
+        - When `keys` is a `list`: Returns a dictionary containing only the selected
+            keys
 
     Example:
         ```python
@@ -687,18 +693,22 @@ class RunnablePick(RunnableSerializable[dict[str, Any], dict[str, Any]]):
             "country": "USA",
         }
 
-        runnable = RunnablePick(keys=["name", "age"])
+        # Single key - returns the value directly
+        runnable_single = RunnablePick(keys="name")
+        result_single = runnable_single.invoke(input_data)
+        print(result_single)  # Output: "John"
 
-        output_data = runnable.invoke(input_data)
-
-        print(output_data)  # Output: {'name': 'John', 'age': 30}
+        # Multiple keys - returns a dictionary
+        runnable_multiple = RunnablePick(keys=["name", "age"])
+        result_multiple = runnable_multiple.invoke(input_data)
+        print(result_multiple)  # Output: {'name': 'John', 'age': 30}
         ```
     """
 
     keys: str | list[str]
 
     def __init__(self, keys: str | list[str], **kwargs: Any) -> None:
-        """Create a RunnablePick.
+        """Create a `RunnablePick`.
 
         Args:
             keys: A single key or a list of keys to pick from the input dictionary.
@@ -708,7 +718,7 @@ class RunnablePick(RunnableSerializable[dict[str, Any], dict[str, Any]]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod

libs/core/langchain_core/runnables/router.py

@@ -40,11 +40,11 @@ class RouterInput(TypedDict):
     key: str
     """The key to route on."""
     input: Any
-    """The input to pass to the selected Runnable."""
+    """The input to pass to the selected `Runnable`."""
 
 
 class RouterRunnable(RunnableSerializable[RouterInput, Output]):
-    """Runnable that routes to a set of Runnables based on Input['key'].
+    """`Runnable` that routes to a set of `Runnable` based on `Input['key']`.
 
     Returns the output of the selected Runnable.
 
@@ -74,10 +74,10 @@ class RouterRunnable(RunnableSerializable[RouterInput, Output]):
         self,
         runnables: Mapping[str, Runnable[Any, Output] | Callable[[Any], Output]],
     ) -> None:
-        """Create a RouterRunnable.
+        """Create a `RouterRunnable`.
 
         Args:
-            runnables: A mapping of keys to Runnables.
+            runnables: A mapping of keys to `Runnable` objects.
         """
         super().__init__(
             runnables={key: coerce_to_runnable(r) for key, r in runnables.items()}
@@ -90,7 +90,7 @@ class RouterRunnable(RunnableSerializable[RouterInput, Output]):
     @classmethod
     @override
     def is_lc_serializable(cls) -> bool:
-        """Return True as this class is serializable."""
+        """Return `True` as this class is serializable."""
         return True
 
     @classmethod

libs/core/langchain_core/runnables/schema.py

@@ -28,7 +28,7 @@ class EventData(TypedDict, total=False):
 
     This field is only available if the `Runnable` raised an exception.
 
-    !!! version-added "Added in version 1.0.0"
+    !!! version-added "Added in `langchain-core` 1.0.0"
     """
     output: Any
     """The output of the `Runnable` that generated the event.

libs/core/langchain_core/runnables/utils.py

@@ -7,8 +7,7 @@ import asyncio
 import inspect
 import sys
 import textwrap
-from collections.abc import Callable, Mapping, Sequence
-from contextvars import Context
+from collections.abc import Mapping, Sequence
 from functools import lru_cache
 from inspect import signature
 from itertools import groupby
@@ -31,9 +30,11 @@ if TYPE_CHECKING:
         AsyncIterable,
         AsyncIterator,
         Awaitable,
+        Callable,
         Coroutine,
         Iterable,
     )
+    from contextvars import Context
 
     from langchain_core.runnables.schema import StreamEvent
 

libs/core/langchain_core/sys_info.py

@@ -125,9 +125,11 @@ def print_sys_info(*, additional_pkgs: Sequence[str] = ()) -> None:
         for dep in sub_dependencies:
             try:
                 dep_version = metadata.version(dep)
-                print(f"> {dep}: {dep_version}")
             except Exception:
-                print(f"> {dep}: Installed. No version info available.")
+                dep_version = None
+
+            if dep_version is not None:
+                print(f"> {dep}: {dep_version}")
 
 
 if __name__ == "__main__":

libs/core/langchain_core/tools/base.py

@@ -386,6 +386,8 @@ class ToolException(Exception):  # noqa: N818
 
 ArgsSchema = TypeBaseModel | dict[str, Any]
 
+_EMPTY_SET: frozenset[str] = frozenset()
+
 
 class BaseTool(RunnableSerializable[str | dict | ToolCall, Any]):
     """Base class for all LangChain tools.
@@ -569,6 +571,11 @@ class ChildTool(BaseTool):
             self.name, full_schema, fields, fn_description=self.description
         )
 
+    @functools.cached_property
+    def _injected_args_keys(self) -> frozenset[str]:
+        # base implementation doesn't manage injected args
+        return _EMPTY_SET
+
     # --- Runnable ---
 
     @override
@@ -649,6 +656,7 @@ class ChildTool(BaseTool):
             if isinstance(input_args, dict):
                 return tool_input
             if issubclass(input_args, BaseModel):
+                # Check args_schema for InjectedToolCallId
                 for k, v in get_all_basemodel_annotations(input_args).items():
                     if _is_injected_arg_type(v, injected_type=InjectedToolCallId):
                         if tool_call_id is None:
@@ -664,6 +672,7 @@ class ChildTool(BaseTool):
                 result = input_args.model_validate(tool_input)
                 result_dict = result.model_dump()
             elif issubclass(input_args, BaseModelV1):
+                # Check args_schema for InjectedToolCallId
                 for k, v in get_all_basemodel_annotations(input_args).items():
                     if _is_injected_arg_type(v, injected_type=InjectedToolCallId):
                         if tool_call_id is None:
@@ -683,9 +692,25 @@ class ChildTool(BaseTool):
                     f"args_schema must be a Pydantic BaseModel, got {self.args_schema}"
                 )
                 raise NotImplementedError(msg)
-            return {
-                k: getattr(result, k) for k, v in result_dict.items() if k in tool_input
+            validated_input = {
+                k: getattr(result, k) for k in result_dict if k in tool_input
             }
+            for k in self._injected_args_keys:
+                if k == "tool_call_id":
+                    if tool_call_id is None:
+                        msg = (
+                            "When tool includes an InjectedToolCallId "
+                            "argument, tool must always be invoked with a full "
+                            "model ToolCall of the form: {'args': {...}, "
+                            "'name': '...', 'type': 'tool_call', "
+                            "'tool_call_id': '...'}"
+                        )
+                        raise ValueError(msg)
+                    validated_input[k] = tool_call_id
+                if k in tool_input:
+                    injected_val = tool_input[k]
+                    validated_input[k] = injected_val
+            return validated_input
         return tool_input
 
     @abstractmethod
@@ -872,16 +897,19 @@ class ChildTool(BaseTool):
                     tool_kwargs |= {config_param: config}
                 response = context.run(self._run, *tool_args, **tool_kwargs)
             if self.response_format == "content_and_artifact":
-                if not isinstance(response, tuple) or len(response) != 2:
-                    msg = (
-                        "Since response_format='content_and_artifact' "
-                        "a two-tuple of the message content and raw tool output is "
-                        f"expected. Instead generated response of type: "
-                        f"{type(response)}."
-                    )
+                msg = (
+                    "Since response_format='content_and_artifact' "
+                    "a two-tuple of the message content and raw tool output is "
+                    f"expected. Instead, generated response is of type: "
+                    f"{type(response)}."
+                )
+                if not isinstance(response, tuple):
                     error_to_raise = ValueError(msg)
                 else:
-                    content, artifact = response
+                    try:
+                        content, artifact = response
+                    except ValueError:
+                        error_to_raise = ValueError(msg)
             else:
                 content = response
         except (ValidationError, ValidationErrorV1) as e:
@@ -998,16 +1026,19 @@ class ChildTool(BaseTool):
                 coro = self._arun(*tool_args, **tool_kwargs)
                 response = await coro_with_context(coro, context)
             if self.response_format == "content_and_artifact":
-                if not isinstance(response, tuple) or len(response) != 2:
-                    msg = (
-                        "Since response_format='content_and_artifact' "
-                        "a two-tuple of the message content and raw tool output is "
-                        f"expected. Instead generated response of type: "
-                        f"{type(response)}."
-                    )
+                msg = (
+                    "Since response_format='content_and_artifact' "
+                    "a two-tuple of the message content and raw tool output is "
+                    f"expected. Instead, generated response is of type: "
+                    f"{type(response)}."
+                )
+                if not isinstance(response, tuple):
                     error_to_raise = ValueError(msg)
                 else:
-                    content, artifact = response
+                    try:
+                        content, artifact = response
+                    except ValueError:
+                        error_to_raise = ValueError(msg)
             else:
                 content = response
         except ValidationError as e:

libs/core/langchain_core/tools/structured.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import functools
 import textwrap
 from collections.abc import Awaitable, Callable
 from inspect import signature
@@ -21,10 +22,12 @@ from langchain_core.callbacks import (
 )
 from langchain_core.runnables import RunnableConfig, run_in_executor
 from langchain_core.tools.base import (
+    _EMPTY_SET,
     FILTERED_ARGS,
     ArgsSchema,
     BaseTool,
     _get_runnable_config_param,
+    _is_injected_arg_type,
     create_schema_from_function,
 )
 from langchain_core.utils.pydantic import is_basemodel_subclass
@@ -241,6 +244,17 @@ class StructuredTool(BaseTool):
             **kwargs,
         )
 
+    @functools.cached_property
+    def _injected_args_keys(self) -> frozenset[str]:
+        fn = self.func or self.coroutine
+        if fn is None:
+            return _EMPTY_SET
+        return frozenset(
+            k
+            for k, v in signature(fn).parameters.items()
+            if _is_injected_arg_type(v.annotation)
+        )
+
 
 def _filter_schema_args(func: Callable) -> list[str]:
     filter_args = list(FILTERED_ARGS)

libs/core/langchain_core/tracers/core.py

@@ -15,12 +15,6 @@ from typing import (
 
 from langchain_core.exceptions import TracerException
 from langchain_core.load import dumpd
-from langchain_core.outputs import (
-    ChatGeneration,
-    ChatGenerationChunk,
-    GenerationChunk,
-    LLMResult,
-)
 from langchain_core.tracers.schemas import Run
 
 if TYPE_CHECKING:
@@ -31,6 +25,12 @@ if TYPE_CHECKING:
 
     from langchain_core.documents import Document
     from langchain_core.messages import BaseMessage
+    from langchain_core.outputs import (
+        ChatGeneration,
+        ChatGenerationChunk,
+        GenerationChunk,
+        LLMResult,
+    )
 
 logger = logging.getLogger(__name__)
 

libs/core/langchain_core/utils/function_calling.py

@@ -8,7 +8,6 @@ import logging
 import types
 import typing
 import uuid
-from collections.abc import Callable
 from typing import (
     TYPE_CHECKING,
     Annotated,
@@ -33,6 +32,8 @@ from langchain_core.utils.json_schema import dereference_refs
 from langchain_core.utils.pydantic import is_basemodel_subclass
 
 if TYPE_CHECKING:
+    from collections.abc import Callable
+
     from langchain_core.tools import BaseTool
 
 logger = logging.getLogger(__name__)
@@ -351,7 +352,8 @@ def convert_to_openai_function(
     Raises:
         ValueError: If function is not in a supported format.
 
-    !!! warning "Behavior changed in 0.3.16"
+    !!! warning "Behavior changed in `langchain-core` 0.3.16"
+
         `description` and `parameters` keys are now optional. Only `name` is
         required and guaranteed to be part of the output.
     """
@@ -412,7 +414,7 @@ def convert_to_openai_function(
     if strict is not None:
         if "strict" in oai_function and oai_function["strict"] != strict:
             msg = (
-                f"Tool/function already has a 'strict' key wth value "
+                f"Tool/function already has a 'strict' key with value "
                 f"{oai_function['strict']} which is different from the explicit "
                 f"`strict` arg received {strict=}."
             )
@@ -475,16 +477,19 @@ def convert_to_openai_tool(
         A dict version of the passed in tool which is compatible with the
         OpenAI tool-calling API.
 
-    !!! warning "Behavior changed in 0.3.16"
+    !!! warning "Behavior changed in `langchain-core` 0.3.16"
+
         `description` and `parameters` keys are now optional. Only `name` is
         required and guaranteed to be part of the output.
 
-    !!! warning "Behavior changed in 0.3.44"
+    !!! warning "Behavior changed in `langchain-core` 0.3.44"
+
         Return OpenAI Responses API-style tools unchanged. This includes
         any dict with `"type"` in `"file_search"`, `"function"`,
         `"computer_use_preview"`, `"web_search_preview"`.
 
-    !!! warning "Behavior changed in 0.3.63"
+    !!! warning "Behavior changed in `langchain-core` 0.3.63"
+
         Added support for OpenAI's image generation built-in tool.
     """
     # Import locally to prevent circular import
@@ -653,6 +658,9 @@ def tool_example_to_messages(
     return messages
 
 
+_MIN_DOCSTRING_BLOCKS = 2
+
+
 def _parse_google_docstring(
     docstring: str | None,
     args: list[str],
@@ -671,7 +679,7 @@ def _parse_google_docstring(
                 arg for arg in args if arg not in {"run_manager", "callbacks", "return"}
             }
             if filtered_annotations and (
-                len(docstring_blocks) < 2
+                len(docstring_blocks) < _MIN_DOCSTRING_BLOCKS
                 or not any(block.startswith("Args:") for block in docstring_blocks[1:])
             ):
                 msg = "Found invalid Google-Style docstring."

libs/core/langchain_core/utils/json.py

@@ -4,11 +4,13 @@ from __future__ import annotations
 
 import json
 import re
-from collections.abc import Callable
-from typing import Any
+from typing import TYPE_CHECKING, Any
 
 from langchain_core.exceptions import OutputParserException
 
+if TYPE_CHECKING:
+    from collections.abc import Callable
+
 
 def _replace_new_line(match: re.Match[str]) -> str:
     value = match.group(2)

libs/core/langchain_core/utils/json_schema.py

@@ -170,28 +170,33 @@ def dereference_refs(
     full_schema: dict | None = None,
     skip_keys: Sequence[str] | None = None,
 ) -> dict:
-    """Resolve and inline JSON Schema $ref references in a schema object.
+    """Resolve and inline JSON Schema `$ref` references in a schema object.
 
-    This function processes a JSON Schema and resolves all $ref references by replacing
-    them with the actual referenced content. It handles both simple references and
-    complex cases like circular references and mixed $ref objects that contain
-    additional properties alongside the $ref.
+    This function processes a JSON Schema and resolves all `$ref` references by
+    replacing them with the actual referenced content.
+
+    Handles both simple references and complex cases like circular references and mixed
+    `$ref` objects that contain additional properties alongside the `$ref`.
 
     Args:
-        schema_obj: The JSON Schema object or fragment to process. This can be a
-            complete schema or just a portion of one.
-        full_schema: The complete schema containing all definitions that $refs might
-            point to. If not provided, defaults to schema_obj (useful when the
-            schema is self-contained).
-        skip_keys: Controls recursion behavior and reference resolution depth:
-            - If `None` (Default): Only recurse under '$defs' and use shallow reference
-              resolution (break cycles but don't deep-inline nested refs)
-            - If provided (even as []): Recurse under all keys and use deep reference
-              resolution (fully inline all nested references)
+        schema_obj: The JSON Schema object or fragment to process.
+
+            This can be a complete schema or just a portion of one.
+        full_schema: The complete schema containing all definitions that `$refs` might
+            point to.
+
+            If not provided, defaults to `schema_obj` (useful when the schema is
+            self-contained).
+        skip_keys: Controls recursion behavior and reference resolution depth.
+
+            - If `None` (Default): Only recurse under `'$defs'` and use shallow
+                reference resolution (break cycles but don't deep-inline nested refs)
+            - If provided (even as `[]`): Recurse under all keys and use deep reference
+                resolution (fully inline all nested references)
 
     Returns:
-        A new dictionary with all $ref references resolved and inlined. The original
-        schema_obj is not modified.
+        A new dictionary with all $ref references resolved and inlined.
+            The original `schema_obj` is not modified.
 
     Examples:
         Basic reference resolution:
@@ -203,7 +208,8 @@ def dereference_refs(
         >>> result = dereference_refs(schema)
         >>> result["properties"]["name"]  # {"type": "string"}
 
-        Mixed $ref with additional properties:
+        Mixed `$ref` with additional properties:
+
         >>> schema = {
         ...     "properties": {
         ...         "name": {"$ref": "#/$defs/base", "description": "User name"}
@@ -215,6 +221,7 @@ def dereference_refs(
         # {"type": "string", "minLength": 1, "description": "User name"}
 
         Handling circular references:
+
         >>> schema = {
         ...     "properties": {"user": {"$ref": "#/$defs/User"}},
         ...     "$defs": {
@@ -227,10 +234,11 @@ def dereference_refs(
         >>> result = dereference_refs(schema)  # Won't cause infinite recursion
 
     !!! note
+
         - Circular references are handled gracefully by breaking cycles
-        - Mixed $ref objects (with both $ref and other properties) are supported
-        - Additional properties in mixed $refs override resolved properties
-        - The $defs section is preserved in the output by default
+        - Mixed `$ref` objects (with both `$ref` and other properties) are supported
+        - Additional properties in mixed `$refs` override resolved properties
+        - The `$defs` section is preserved in the output by default
     """
     full = full_schema or schema_obj
     keys_to_skip = list(skip_keys) if skip_keys is not None else ["$defs"]

libs/core/langchain_core/utils/mustache.py

@@ -374,15 +374,29 @@ def _get_key(
                 if resolved_scope in (0, False):
                     return resolved_scope
                 # Move into the scope
-                try:
-                    # Try subscripting (Normal dictionaries)
-                    resolved_scope = cast("dict[str, Any]", resolved_scope)[child]
-                except (TypeError, AttributeError):
+                if isinstance(resolved_scope, dict):
                     try:
-                        resolved_scope = getattr(resolved_scope, child)
-                    except (TypeError, AttributeError):
-                        # Try as a list
-                        resolved_scope = resolved_scope[int(child)]  # type: ignore[index]
+                        resolved_scope = resolved_scope[child]
+                    except (KeyError, TypeError):
+                        # Key not found - will be caught by outer try-except
+                        msg = f"Key {child!r} not found in dict"
+                        raise KeyError(msg) from None
+                elif isinstance(resolved_scope, (list, tuple)):
+                    try:
+                        resolved_scope = resolved_scope[int(child)]
+                    except (ValueError, IndexError, TypeError):
+                        # Invalid index - will be caught by outer try-except
+                        msg = f"Invalid index {child!r} for list/tuple"
+                        raise IndexError(msg) from None
+                else:
+                    # Reject everything else for security
+                    # This prevents traversing into arbitrary Python objects
+                    msg = (
+                        f"Cannot traverse into {type(resolved_scope).__name__}. "
+                        "Mustache templates only support dict, list, and tuple. "
+                        f"Got: {type(resolved_scope)}"
+                    )
+                    raise TypeError(msg)  # noqa: TRY301
 
             try:
                 # This allows for custom falsy data types
@@ -393,8 +407,9 @@ def _get_key(
                 if resolved_scope in (0, False):
                     return resolved_scope
                 return resolved_scope or ""
-        except (AttributeError, KeyError, IndexError, ValueError):
+        except (AttributeError, KeyError, IndexError, ValueError, TypeError):
             # We couldn't find the key in the current scope
+            # TypeError: Attempted to traverse into non-dict/list type
             # We'll try again on the next pass
             pass
 

libs/core/langchain_core/utils/pydantic.py

@@ -5,7 +5,6 @@ from __future__ import annotations
 import inspect
 import textwrap
 import warnings
-from collections.abc import Callable
 from contextlib import nullcontext
 from functools import lru_cache, wraps
 from types import GenericAlias
@@ -41,10 +40,12 @@ from pydantic.json_schema import (
 )
 from pydantic.v1 import BaseModel as BaseModelV1
 from pydantic.v1 import create_model as create_model_v1
-from pydantic.v1.fields import ModelField
 from typing_extensions import deprecated, override
 
 if TYPE_CHECKING:
+    from collections.abc import Callable
+
+    from pydantic.v1.fields import ModelField
     from pydantic_core import core_schema
 
 PYDANTIC_VERSION = version.parse(pydantic.__version__)
@@ -65,8 +66,8 @@ def get_pydantic_major_version() -> int:
 PYDANTIC_MAJOR_VERSION = PYDANTIC_VERSION.major
 PYDANTIC_MINOR_VERSION = PYDANTIC_VERSION.minor
 
-IS_PYDANTIC_V1 = PYDANTIC_VERSION.major == 1
-IS_PYDANTIC_V2 = PYDANTIC_VERSION.major == 2
+IS_PYDANTIC_V1 = False
+IS_PYDANTIC_V2 = True
 
 PydanticBaseModel = BaseModel
 TypeBaseModel = type[BaseModel]

libs/core/langchain_core/vectorstores/base.py

@@ -11,7 +11,6 @@ import logging
 import math
 import warnings
 from abc import ABC, abstractmethod
-from collections.abc import Callable
 from itertools import cycle
 from typing import (
     TYPE_CHECKING,
@@ -29,7 +28,7 @@ from langchain_core.retrievers import BaseRetriever, LangSmithRetrieverParams
 from langchain_core.runnables.config import run_in_executor
 
 if TYPE_CHECKING:
-    from collections.abc import Collection, Iterable, Iterator, Sequence
+    from collections.abc import Callable, Collection, Iterable, Iterator, Sequence
 
     from langchain_core.callbacks.manager import (
         AsyncCallbackManagerForRetrieverRun,
@@ -295,8 +294,9 @@ class VectorStore(ABC):
 
         Args:
             query: Input text.
-            search_type: Type of search to perform. Can be `'similarity'`, `'mmr'`, or
-                `'similarity_score_threshold'`.
+            search_type: Type of search to perform.
+
+                Can be `'similarity'`, `'mmr'`, or `'similarity_score_threshold'`.
             **kwargs: Arguments to pass to the search method.
 
         Returns:
@@ -329,8 +329,9 @@ class VectorStore(ABC):
 
         Args:
             query: Input text.
-            search_type: Type of search to perform. Can be `'similarity'`, `'mmr'`, or
-                `'similarity_score_threshold'`.
+            search_type: Type of search to perform.
+
+                Can be `'similarity'`, `'mmr'`, or `'similarity_score_threshold'`.
             **kwargs: Arguments to pass to the search method.
 
         Returns:
@@ -461,9 +462,10 @@ class VectorStore(ABC):
         Args:
             query: Input text.
             k: Number of `Document` objects to return.
-            **kwargs: kwargs to be passed to similarity search. Should include
-                `score_threshold`, An optional floating point value between `0` to `1`
-                to filter the resulting set of retrieved docs
+            **kwargs: Kwargs to be passed to similarity search.
+
+                Should include `score_threshold`, an optional floating point value
+                between `0` to `1` to filter the resulting set of retrieved docs.
 
         Returns:
             List of tuples of `(doc, similarity_score)`
@@ -488,9 +490,10 @@ class VectorStore(ABC):
         Args:
             query: Input text.
             k: Number of `Document` objects to return.
-            **kwargs: kwargs to be passed to similarity search. Should include
-                `score_threshold`, An optional floating point value between `0` to `1`
-                to filter the resulting set of retrieved docs
+            **kwargs: Kwargs to be passed to similarity search.
+
+                Should include `score_threshold`, an optional floating point value
+                between `0` to `1` to filter the resulting set of retrieved docs.
 
         Returns:
             List of tuples of `(doc, similarity_score)`
@@ -512,9 +515,10 @@ class VectorStore(ABC):
         Args:
             query: Input text.
             k: Number of `Document` objects to return.
-            **kwargs: kwargs to be passed to similarity search. Should include
-                `score_threshold`, An optional floating point value between `0` to `1`
-                to filter the resulting set of retrieved docs
+            **kwargs: Kwargs to be passed to similarity search.
+
+                Should include `score_threshold`, an optional floating point value
+                between `0` to `1` to filter the resulting set of retrieved docs.
 
         Returns:
             List of tuples of `(doc, similarity_score)`.
@@ -561,9 +565,10 @@ class VectorStore(ABC):
         Args:
             query: Input text.
             k: Number of `Document` objects to return.
-            **kwargs: kwargs to be passed to similarity search. Should include
-                `score_threshold`, An optional floating point value between `0` to `1`
-                to filter the resulting set of retrieved docs
+            **kwargs: Kwargs to be passed to similarity search.
+
+                Should include `score_threshold`, an optional floating point value
+                between `0` to `1` to filter the resulting set of retrieved docs.
 
         Returns:
             List of tuples of `(doc, similarity_score)`
@@ -901,13 +906,15 @@ class VectorStore(ABC):
 
         Args:
             **kwargs: Keyword arguments to pass to the search function.
+
                 Can include:
 
                 * `search_type`: Defines the type of search that the Retriever should
                     perform. Can be `'similarity'` (default), `'mmr'`, or
                     `'similarity_score_threshold'`.
-                * `search_kwargs`: Keyword arguments to pass to the search function. Can
-                    include things like:
+                * `search_kwargs`: Keyword arguments to pass to the search function.
+
+                    Can include things like:
 
                     * `k`: Amount of documents to return (Default: `4`)
                     * `score_threshold`: Minimum relevance threshold

libs/core/langchain_core/vectorstores/in_memory.py

@@ -4,7 +4,6 @@ from __future__ import annotations
 
 import json
 import uuid
-from collections.abc import Callable
 from pathlib import Path
 from typing import (
     TYPE_CHECKING,
@@ -20,7 +19,7 @@ from langchain_core.vectorstores.utils import _cosine_similarity as cosine_simil
 from langchain_core.vectorstores.utils import maximal_marginal_relevance
 
 if TYPE_CHECKING:
-    from collections.abc import Iterator, Sequence
+    from collections.abc import Callable, Iterator, Sequence
 
     from langchain_core.embeddings import Embeddings
 

libs/core/langchain_core/version.py

@@ -1,3 +1,3 @@
 """langchain-core version information and utilities."""
 
-VERSION = "1.0.3"
+VERSION = "1.1.0"

libs/core/pyproject.toml

@@ -3,8 +3,13 @@ requires = ["hatchling"]
 build-backend = "hatchling.build"
 
 [project]
-authors = []
+name = "langchain-core"
+description = "Building applications with LLMs through composability"
 license = {text = "MIT"}
+readme = "README.md"
+authors = []
+
+version = "1.1.0"
 requires-python = ">=3.10.0,<4.0.0"
 dependencies = [
     "langsmith>=0.3.45,<1.0.0",
@@ -15,10 +20,6 @@ dependencies = [
     "packaging>=23.2.0,<26.0.0",
     "pydantic>=2.7.4,<3.0.0",
 ]
-name = "langchain-core"
-version = "1.0.3"
-description = "Building applications with LLMs through composability"
-readme = "README.md"
 
 [project.urls]
 Homepage = "https://docs.langchain.com/"
@@ -35,7 +36,6 @@ typing = [
     "mypy>=1.18.1,<1.19.0",
     "types-pyyaml>=6.0.12.2,<7.0.0.0",
     "types-requests>=2.28.11.5,<3.0.0.0",
-    "langchain-model-profiles",
     "langchain-text-splitters",
 ]
 dev = [
@@ -57,7 +57,6 @@ test = [
     "blockbuster>=1.5.18,<1.6.0",
     "numpy>=1.26.4; python_version<'3.13'",
     "numpy>=2.1.0; python_version>='3.13'",
-    "langchain-model-profiles",
     "langchain-tests",
     "pytest-benchmark",
     "pytest-codspeed",
@@ -65,7 +64,6 @@ test = [
 test_integration = []
 
 [tool.uv.sources]
-langchain-model-profiles = { path = "../model-profiles" }
 langchain-tests = { path = "../standard-tests" }
 langchain-text-splitters = { path = "../text-splitters" }
 
@@ -104,7 +102,6 @@ ignore = [
     "ANN401",  # No Any types
     "BLE",     # Blind exceptions
     "ERA",     # No commented-out code
-    "PLR2004", # Comparison to magic number
 ]
 unfixable = [
     "B028",    # People should intentionally tune the stacklevel
@@ -125,7 +122,7 @@ ignore-var-parameters = true  # ignore missing documentation for *args and **kwa
 "langchain_core/utils/mustache.py" = [ "PLW0603",]
 "langchain_core/sys_info.py" = [ "T201",]
 "tests/unit_tests/test_tools.py" = [ "ARG",]
-"tests/**" = [ "D1", "S", "SLF",]
+"tests/**" = [ "D1", "PLR2004", "S", "SLF",]
 "scripts/**" = [ "INP", "S",]
 
 [tool.coverage.run]
@@ -133,7 +130,10 @@ omit = [ "tests/*",]
 
 [tool.pytest.ini_options]
 addopts = "--snapshot-warn-unused --strict-markers --strict-config --durations=5"
-markers = [ "requires: mark tests as requiring a specific library", "compile: mark placeholder test used to compile integration tests without running them", ]
+markers = [
+    "requires: mark tests as requiring a specific library",
+    "compile: mark placeholder test used to compile integration tests without running them",
+]
 asyncio_mode = "auto"
-filterwarnings = [ "ignore::langchain_core._api.beta_decorator.LangChainBetaWarning",]
 asyncio_default_fixture_loop_scope = "function"
+filterwarnings = [ "ignore::langchain_core._api.beta_decorator.LangChainBetaWarning",]

libs/core/tests/unit_tests/callbacks/test_async_callback_manager.py

@@ -148,4 +148,65 @@ async def test_inline_handlers_share_parent_context_multiple() -> None:
             2,
             3,
             3,
-        ], f"Expected order of states was broken due to context loss. Got {states}"
+        ]
+
+
+async def test_shielded_callback_context_preservation() -> None:
+    """Verify that shielded callbacks preserve context variables.
+
+    This test specifically addresses the issue where async callbacks decorated
+    with @shielded do not properly preserve context variables, breaking
+    instrumentation and other context-dependent functionality.
+
+    The issue manifests in callbacks that use the @shielded decorator:
+    * on_llm_end
+    * on_llm_error
+    * on_chain_end
+    * on_chain_error
+    * And other shielded callback methods
+    """
+    context_var: contextvars.ContextVar[str] = contextvars.ContextVar("test_context")
+
+    class ContextTestHandler(AsyncCallbackHandler):
+        """Handler that reads context variables in shielded callbacks."""
+
+        def __init__(self) -> None:
+            self.run_inline = False
+            self.context_values: list[str] = []
+
+        @override
+        async def on_llm_end(self, response: Any, **kwargs: Any) -> None:
+            """This method is decorated with @shielded in the run manager."""
+            # This should preserve the context variable value
+            self.context_values.append(context_var.get("not_found"))
+
+        @override
+        async def on_chain_end(self, outputs: Any, **kwargs: Any) -> None:
+            """This method is decorated with @shielded in the run manager."""
+            # This should preserve the context variable value
+            self.context_values.append(context_var.get("not_found"))
+
+    # Set up the test context
+    context_var.set("test_value")
+    handler = ContextTestHandler()
+    manager = AsyncCallbackManager(handlers=[handler])
+
+    # Create run managers that have the shielded methods
+    llm_managers = await manager.on_llm_start({}, ["test prompt"])
+    llm_run_manager = llm_managers[0]
+
+    chain_run_manager = await manager.on_chain_start({}, {"test": "input"})
+
+    # Test LLM end callback (which is shielded)
+    await llm_run_manager.on_llm_end({"response": "test"})  # type: ignore[arg-type]
+
+    # Test Chain end callback (which is shielded)
+    await chain_run_manager.on_chain_end({"output": "test"})
+
+    # The context should be preserved in shielded callbacks
+    # This was the main issue - shielded decorators were not preserving context
+    assert handler.context_values == ["test_value", "test_value"], (
+        f"Expected context values ['test_value', 'test_value'], "
+        f"but got {handler.context_values}. "
+        f"This indicates the shielded decorator is not preserving context variables."
+    )

libs/core/tests/unit_tests/indexing/test_hashed_document.py

@@ -33,7 +33,7 @@ def test_hashing() -> None:
     # hash should be deterministic
     assert hashed_document.id == "fd1dc827-051b-537d-a1fe-1fa043e8b276"
 
-    # Verify that hashing with sha1 is determinstic
+    # Verify that hashing with sha1 is deterministic
     another_hashed_document = _get_document_with_hash(document, key_encoder="sha1")
     assert another_hashed_document.id == hashed_document.id
 

libs/core/tests/unit_tests/language_models/chat_models/test_base.py

@@ -18,6 +18,7 @@ from langchain_core.language_models import (
     ParrotFakeChatModel,
 )
 from langchain_core.language_models._utils import _normalize_messages
+from langchain_core.language_models.chat_models import _generate_response_from_error
 from langchain_core.language_models.fake_chat_models import (
     FakeListChatModelError,
     GenericFakeChatModel,
@@ -1221,16 +1222,99 @@ def test_get_ls_params() -> None:
 
 def test_model_profiles() -> None:
     model = GenericFakeChatModel(messages=iter([]))
-    profile = model.profile
-    assert profile == {}
+    assert model.profile is None
 
-    class MyModel(GenericFakeChatModel):
-        model: str = "gpt-5"
+    model_with_profile = GenericFakeChatModel(
+        messages=iter([]), profile={"max_input_tokens": 100}
+    )
+    assert model_with_profile.profile == {"max_input_tokens": 100}
 
-        @property
-        def _llm_type(self) -> str:
-            return "openai-chat"
 
-    model = MyModel(messages=iter([]))
-    profile = model.profile
-    assert profile
+class MockResponse:
+    """Mock response for testing _generate_response_from_error."""
+
+    def __init__(
+        self,
+        status_code: int = 400,
+        headers: dict[str, str] | None = None,
+        json_data: dict[str, Any] | None = None,
+        json_raises: type[Exception] | None = None,
+        text_raises: type[Exception] | None = None,
+    ):
+        self.status_code = status_code
+        self.headers = headers or {}
+        self._json_data = json_data
+        self._json_raises = json_raises
+        self._text_raises = text_raises
+
+    def json(self) -> dict[str, Any]:
+        if self._json_raises:
+            msg = "JSON parsing failed"
+            raise self._json_raises(msg)
+        return self._json_data or {}
+
+    @property
+    def text(self) -> str:
+        if self._text_raises:
+            msg = "Text access failed"
+            raise self._text_raises(msg)
+        return ""
+
+
+class MockAPIError(Exception):
+    """Mock API error with response attribute."""
+
+    def __init__(self, message: str, response: MockResponse | None = None):
+        super().__init__(message)
+        self.message = message
+        if response is not None:
+            self.response = response
+
+
+def test_generate_response_from_error_with_valid_json() -> None:
+    """Test `_generate_response_from_error` with valid JSON response."""
+    response = MockResponse(
+        status_code=400,
+        headers={"content-type": "application/json"},
+        json_data={"error": {"message": "Bad request", "type": "invalid_request"}},
+    )
+    error = MockAPIError("API Error", response=response)
+
+    generations = _generate_response_from_error(error)
+
+    assert len(generations) == 1
+    generation = generations[0]
+    assert isinstance(generation, ChatGeneration)
+    assert isinstance(generation.message, AIMessage)
+    assert generation.message.content == ""
+
+    metadata = generation.message.response_metadata
+    assert metadata["body"] == {
+        "error": {"message": "Bad request", "type": "invalid_request"}
+    }
+    assert metadata["headers"] == {"content-type": "application/json"}
+    assert metadata["status_code"] == 400
+
+
+def test_generate_response_from_error_handles_streaming_response_failure() -> None:
+    # Simulates scenario where accessing response.json() or response.text
+    # raises ResponseNotRead on streaming responses
+    response = MockResponse(
+        status_code=400,
+        headers={"content-type": "application/json"},
+        json_raises=Exception,  # Simulates ResponseNotRead or similar
+        text_raises=Exception,
+    )
+    error = MockAPIError("API Error", response=response)
+
+    # This should NOT raise an exception, but should handle it gracefully
+    generations = _generate_response_from_error(error)
+
+    assert len(generations) == 1
+    generation = generations[0]
+    metadata = generation.message.response_metadata
+
+    # When both fail, body should be None instead of raising an exception
+    assert metadata["body"] is None
+    assert metadata["headers"] == {"content-type": "application/json"}
+    assert metadata["status_code"] == 400

libs/core/tests/unit_tests/language_models/test_imports.py

@@ -18,6 +18,8 @@ EXPECTED_ALL = [
     "FakeStreamingListLLM",
     "FakeListLLM",
     "ParrotFakeChatModel",
+    "ModelProfile",
+    "ModelProfileRegistry",
     "is_openai_data_block",
 ]
 

libs/core/tests/unit_tests/messages/block_translators/test_groq.py

@@ -0,0 +1,140 @@
+"""Test groq block translator."""
+
+from typing import cast
+
+import pytest
+
+from langchain_core.messages import AIMessage
+from langchain_core.messages import content as types
+from langchain_core.messages.base import _extract_reasoning_from_additional_kwargs
+from langchain_core.messages.block_translators import PROVIDER_TRANSLATORS
+from langchain_core.messages.block_translators.groq import (
+    _parse_code_json,
+    translate_content,
+)
+
+
+def test_groq_translator_registered() -> None:
+    """Test that groq translator is properly registered."""
+    assert "groq" in PROVIDER_TRANSLATORS
+    assert "translate_content" in PROVIDER_TRANSLATORS["groq"]
+    assert "translate_content_chunk" in PROVIDER_TRANSLATORS["groq"]
+
+
+def test_extract_reasoning_from_additional_kwargs_exists() -> None:
+    """Test that _extract_reasoning_from_additional_kwargs can be imported."""
+    # Verify it's callable
+    assert callable(_extract_reasoning_from_additional_kwargs)
+
+
+def test_groq_translate_content_basic() -> None:
+    """Test basic groq content translation."""
+    # Test with simple text message
+    message = AIMessage(content="Hello world")
+    blocks = translate_content(message)
+
+    assert isinstance(blocks, list)
+    assert len(blocks) == 1
+    assert blocks[0]["type"] == "text"
+    assert blocks[0]["text"] == "Hello world"
+
+
+def test_groq_translate_content_with_reasoning() -> None:
+    """Test groq content translation with reasoning content."""
+    # Test with reasoning content in additional_kwargs
+    message = AIMessage(
+        content="Final answer",
+        additional_kwargs={"reasoning_content": "Let me think about this..."},
+    )
+    blocks = translate_content(message)
+
+    assert isinstance(blocks, list)
+    assert len(blocks) == 2
+
+    # First block should be reasoning
+    assert blocks[0]["type"] == "reasoning"
+    assert blocks[0]["reasoning"] == "Let me think about this..."
+
+    # Second block should be text
+    assert blocks[1]["type"] == "text"
+    assert blocks[1]["text"] == "Final answer"
+
+
+def test_groq_translate_content_with_tool_calls() -> None:
+    """Test groq content translation with tool calls."""
+    # Test with tool calls
+    message = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "name": "search",
+                "args": {"query": "test"},
+                "id": "call_123",
+            }
+        ],
+    )
+    blocks = translate_content(message)
+
+    assert isinstance(blocks, list)
+    assert len(blocks) == 1
+    assert blocks[0]["type"] == "tool_call"
+    assert blocks[0]["name"] == "search"
+    assert blocks[0]["args"] == {"query": "test"}
+    assert blocks[0]["id"] == "call_123"
+
+
+def test_groq_translate_content_with_executed_tools() -> None:
+    """Test groq content translation with executed tools (built-in tools)."""
+    # Test with executed_tools in additional_kwargs (Groq built-in tools)
+    message = AIMessage(
+        content="",
+        additional_kwargs={
+            "executed_tools": [
+                {
+                    "type": "python",
+                    "arguments": '{"code": "print(\\"hello\\")"}',
+                    "output": "hello\\n",
+                }
+            ]
+        },
+    )
+    blocks = translate_content(message)
+
+    assert isinstance(blocks, list)
+    # Should have server_tool_call and server_tool_result
+    assert len(blocks) >= 2
+
+    # Check for server_tool_call
+    tool_call_blocks = [
+        cast("types.ServerToolCall", b)
+        for b in blocks
+        if b.get("type") == "server_tool_call"
+    ]
+    assert len(tool_call_blocks) == 1
+    assert tool_call_blocks[0]["name"] == "code_interpreter"
+    assert "code" in tool_call_blocks[0]["args"]
+
+    # Check for server_tool_result
+    tool_result_blocks = [
+        cast("types.ServerToolResult", b)
+        for b in blocks
+        if b.get("type") == "server_tool_result"
+    ]
+    assert len(tool_result_blocks) == 1
+    assert tool_result_blocks[0]["output"] == "hello\\n"
+    assert tool_result_blocks[0]["status"] == "success"
+
+
+def test_parse_code_json() -> None:
+    """Test the _parse_code_json helper function."""
+    # Test valid code JSON
+    result = _parse_code_json('{"code": "print(\'hello\')"}')
+    assert result == {"code": "print('hello')"}
+
+    # Test code with unescaped quotes (Groq format)
+    result = _parse_code_json('{"code": "print("hello")"}')
+    assert result == {"code": 'print("hello")'}
+
+    # Test invalid format raises ValueError
+    with pytest.raises(ValueError, match="Could not extract Python code"):
+        _parse_code_json('{"invalid": "format"}')

libs/core/tests/unit_tests/output_parsers/test_openai_tools.py

@@ -1,3 +1,4 @@
+import sys
 from collections.abc import AsyncIterator, Iterator
 from typing import Any
 
@@ -886,3 +887,461 @@ def test_max_tokens_error(caplog: Any) -> None:
         "`max_tokens` stop reason" in msg and record.levelname == "ERROR"
         for record, msg in zip(caplog.records, caplog.messages, strict=False)
     )
+
+
+def test_pydantic_tools_parser_with_mixed_pydantic_versions() -> None:
+    """Test PydanticToolsParser with both Pydantic v1 and v2 models."""
+    # For Python 3.14+ compatibility, use create_model for Pydantic v1
+    if sys.version_info >= (3, 14):
+        WeatherV1 = pydantic.v1.create_model(  # noqa: N806
+            "WeatherV1",
+            __doc__="Weather information using Pydantic v1.",
+            temperature=(int, ...),
+            conditions=(str, ...),
+        )
+    else:
+
+        class WeatherV1(pydantic.v1.BaseModel):
+            """Weather information using Pydantic v1."""
+
+            temperature: int
+            conditions: str
+
+    class LocationV2(BaseModel):
+        """Location information using Pydantic v2."""
+
+        city: str
+        country: str
+
+    # Test with Pydantic v1 model
+    parser_v1 = PydanticToolsParser(tools=[WeatherV1])
+    message_v1 = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_weather",
+                "name": "WeatherV1",
+                "args": {"temperature": 25, "conditions": "sunny"},
+            }
+        ],
+    )
+    generation_v1 = ChatGeneration(message=message_v1)
+    result_v1 = parser_v1.parse_result([generation_v1])
+
+    assert len(result_v1) == 1
+    assert isinstance(result_v1[0], WeatherV1)
+    assert result_v1[0].temperature == 25  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1[0].conditions == "sunny"  # type: ignore[attr-defined,unused-ignore]
+
+    # Test with Pydantic v2 model
+    parser_v2 = PydanticToolsParser(tools=[LocationV2])
+    message_v2 = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_location",
+                "name": "LocationV2",
+                "args": {"city": "Paris", "country": "France"},
+            }
+        ],
+    )
+    generation_v2 = ChatGeneration(message=message_v2)
+    result_v2 = parser_v2.parse_result([generation_v2])
+
+    assert len(result_v2) == 1
+    assert isinstance(result_v2[0], LocationV2)
+    assert result_v2[0].city == "Paris"
+    assert result_v2[0].country == "France"
+
+    # Test with both v1 and v2 models
+    parser_mixed = PydanticToolsParser(tools=[WeatherV1, LocationV2])
+    message_mixed = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_weather",
+                "name": "WeatherV1",
+                "args": {"temperature": 20, "conditions": "cloudy"},
+            },
+            {
+                "id": "call_location",
+                "name": "LocationV2",
+                "args": {"city": "London", "country": "UK"},
+            },
+        ],
+    )
+    generation_mixed = ChatGeneration(message=message_mixed)
+    result_mixed = parser_mixed.parse_result([generation_mixed])
+
+    assert len(result_mixed) == 2
+    assert isinstance(result_mixed[0], WeatherV1)
+    assert result_mixed[0].temperature == 20  # type: ignore[attr-defined,unused-ignore]
+    assert isinstance(result_mixed[1], LocationV2)
+    assert result_mixed[1].city == "London"
+
+
+def test_pydantic_tools_parser_with_custom_title() -> None:
+    """Test PydanticToolsParser with Pydantic v2 model using custom title."""
+
+    class CustomTitleTool(BaseModel):
+        """Tool with custom title in model config."""
+
+        model_config = {"title": "MyCustomToolName"}
+
+        value: int
+        description: str
+
+    # Test with custom title - tool should be callable by custom name
+    parser = PydanticToolsParser(tools=[CustomTitleTool])
+    message = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_custom",
+                "name": "MyCustomToolName",
+                "args": {"value": 42, "description": "test"},
+            }
+        ],
+    )
+    generation = ChatGeneration(message=message)
+    result = parser.parse_result([generation])
+
+    assert len(result) == 1
+    assert isinstance(result[0], CustomTitleTool)
+    assert result[0].value == 42
+    assert result[0].description == "test"
+
+
+def test_pydantic_tools_parser_name_dict_fallback() -> None:
+    """Test that name_dict properly falls back to __name__ when title is None."""
+
+    class ToolWithoutTitle(BaseModel):
+        """Tool without explicit title."""
+
+        data: str
+
+    # Ensure model_config doesn't have a title or it's None
+    # (This is the default behavior)
+    parser = PydanticToolsParser(tools=[ToolWithoutTitle])
+    message = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_no_title",
+                "name": "ToolWithoutTitle",
+                "args": {"data": "test_data"},
+            }
+        ],
+    )
+    generation = ChatGeneration(message=message)
+    result = parser.parse_result([generation])
+
+    assert len(result) == 1
+    assert isinstance(result[0], ToolWithoutTitle)
+    assert result[0].data == "test_data"
+
+
+def test_pydantic_tools_parser_with_nested_models() -> None:
+    """Test PydanticToolsParser with nested Pydantic v1 and v2 models."""
+    # Nested v1 models
+    if sys.version_info >= (3, 14):
+        AddressV1 = pydantic.v1.create_model(  # noqa: N806
+            "AddressV1",
+            __doc__="Address using Pydantic v1.",
+            street=(str, ...),
+            city=(str, ...),
+            zip_code=(str, ...),
+        )
+        PersonV1 = pydantic.v1.create_model(  # noqa: N806
+            "PersonV1",
+            __doc__="Person with nested address using Pydantic v1.",
+            name=(str, ...),
+            age=(int, ...),
+            address=(AddressV1, ...),
+        )
+    else:
+
+        class AddressV1(pydantic.v1.BaseModel):
+            """Address using Pydantic v1."""
+
+            street: str
+            city: str
+            zip_code: str
+
+        class PersonV1(pydantic.v1.BaseModel):
+            """Person with nested address using Pydantic v1."""
+
+            name: str
+            age: int
+            address: AddressV1
+
+    # Nested v2 models
+    class CoordinatesV2(BaseModel):
+        """Coordinates using Pydantic v2."""
+
+        latitude: float
+        longitude: float
+
+    class LocationV2(BaseModel):
+        """Location with nested coordinates using Pydantic v2."""
+
+        name: str
+        coordinates: CoordinatesV2
+
+    # Test with nested Pydantic v1 model
+    parser_v1 = PydanticToolsParser(tools=[PersonV1])
+    message_v1 = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_person",
+                "name": "PersonV1",
+                "args": {
+                    "name": "Alice",
+                    "age": 30,
+                    "address": {
+                        "street": "123 Main St",
+                        "city": "Springfield",
+                        "zip_code": "12345",
+                    },
+                },
+            }
+        ],
+    )
+    generation_v1 = ChatGeneration(message=message_v1)
+    result_v1 = parser_v1.parse_result([generation_v1])
+
+    assert len(result_v1) == 1
+    assert isinstance(result_v1[0], PersonV1)
+    assert result_v1[0].name == "Alice"  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1[0].age == 30  # type: ignore[attr-defined,unused-ignore]
+    assert isinstance(result_v1[0].address, AddressV1)  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1[0].address.street == "123 Main St"  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1[0].address.city == "Springfield"  # type: ignore[attr-defined,unused-ignore]
+
+    # Test with nested Pydantic v2 model
+    parser_v2 = PydanticToolsParser(tools=[LocationV2])
+    message_v2 = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_location",
+                "name": "LocationV2",
+                "args": {
+                    "name": "Eiffel Tower",
+                    "coordinates": {"latitude": 48.8584, "longitude": 2.2945},
+                },
+            }
+        ],
+    )
+    generation_v2 = ChatGeneration(message=message_v2)
+    result_v2 = parser_v2.parse_result([generation_v2])
+
+    assert len(result_v2) == 1
+    assert isinstance(result_v2[0], LocationV2)
+    assert result_v2[0].name == "Eiffel Tower"
+    assert isinstance(result_v2[0].coordinates, CoordinatesV2)
+    assert result_v2[0].coordinates.latitude == 48.8584
+    assert result_v2[0].coordinates.longitude == 2.2945
+
+    # Test with both nested models in one message
+    parser_mixed = PydanticToolsParser(tools=[PersonV1, LocationV2])
+    message_mixed = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_person",
+                "name": "PersonV1",
+                "args": {
+                    "name": "Bob",
+                    "age": 25,
+                    "address": {
+                        "street": "456 Oak Ave",
+                        "city": "Portland",
+                        "zip_code": "97201",
+                    },
+                },
+            },
+            {
+                "id": "call_location",
+                "name": "LocationV2",
+                "args": {
+                    "name": "Golden Gate Bridge",
+                    "coordinates": {"latitude": 37.8199, "longitude": -122.4783},
+                },
+            },
+        ],
+    )
+    generation_mixed = ChatGeneration(message=message_mixed)
+    result_mixed = parser_mixed.parse_result([generation_mixed])
+
+    assert len(result_mixed) == 2
+    assert isinstance(result_mixed[0], PersonV1)
+    assert result_mixed[0].name == "Bob"  # type: ignore[attr-defined,unused-ignore]
+    assert result_mixed[0].address.city == "Portland"  # type: ignore[attr-defined,unused-ignore]
+    assert isinstance(result_mixed[1], LocationV2)
+    assert result_mixed[1].name == "Golden Gate Bridge"
+    assert result_mixed[1].coordinates.latitude == 37.8199
+
+
+def test_pydantic_tools_parser_with_optional_fields() -> None:
+    """Test PydanticToolsParser with optional fields in v1 and v2 models."""
+    if sys.version_info >= (3, 14):
+        ProductV1 = pydantic.v1.create_model(  # noqa: N806
+            "ProductV1",
+            __doc__="Product with optional fields using Pydantic v1.",
+            name=(str, ...),
+            price=(float, ...),
+            description=(str | None, None),
+            stock=(int, 0),
+        )
+    else:
+
+        class ProductV1(pydantic.v1.BaseModel):
+            """Product with optional fields using Pydantic v1."""
+
+            name: str
+            price: float
+            description: str | None = None
+            stock: int = 0
+
+    # v2 model with optional fields
+    class UserV2(BaseModel):
+        """User with optional fields using Pydantic v2."""
+
+        username: str
+        email: str
+        bio: str | None = None
+        age: int | None = None
+
+    # Test v1 with all fields provided
+    parser_v1_full = PydanticToolsParser(tools=[ProductV1])
+    message_v1_full = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_product_full",
+                "name": "ProductV1",
+                "args": {
+                    "name": "Laptop",
+                    "price": 999.99,
+                    "description": "High-end laptop",
+                    "stock": 50,
+                },
+            }
+        ],
+    )
+    generation_v1_full = ChatGeneration(message=message_v1_full)
+    result_v1_full = parser_v1_full.parse_result([generation_v1_full])
+
+    assert len(result_v1_full) == 1
+    assert isinstance(result_v1_full[0], ProductV1)
+    assert result_v1_full[0].name == "Laptop"  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1_full[0].price == 999.99  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1_full[0].description == "High-end laptop"  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1_full[0].stock == 50  # type: ignore[attr-defined,unused-ignore]
+
+    # Test v1 with only required fields
+    parser_v1_minimal = PydanticToolsParser(tools=[ProductV1])
+    message_v1_minimal = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_product_minimal",
+                "name": "ProductV1",
+                "args": {"name": "Mouse", "price": 29.99},
+            }
+        ],
+    )
+    generation_v1_minimal = ChatGeneration(message=message_v1_minimal)
+    result_v1_minimal = parser_v1_minimal.parse_result([generation_v1_minimal])
+
+    assert len(result_v1_minimal) == 1
+    assert isinstance(result_v1_minimal[0], ProductV1)
+    assert result_v1_minimal[0].name == "Mouse"  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1_minimal[0].price == 29.99  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1_minimal[0].description is None  # type: ignore[attr-defined,unused-ignore]
+    assert result_v1_minimal[0].stock == 0  # type: ignore[attr-defined,unused-ignore]
+
+    # Test v2 with all fields provided
+    parser_v2_full = PydanticToolsParser(tools=[UserV2])
+    message_v2_full = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_user_full",
+                "name": "UserV2",
+                "args": {
+                    "username": "john_doe",
+                    "email": "john@example.com",
+                    "bio": "Software developer",
+                    "age": 28,
+                },
+            }
+        ],
+    )
+    generation_v2_full = ChatGeneration(message=message_v2_full)
+    result_v2_full = parser_v2_full.parse_result([generation_v2_full])
+
+    assert len(result_v2_full) == 1
+    assert isinstance(result_v2_full[0], UserV2)
+    assert result_v2_full[0].username == "john_doe"
+    assert result_v2_full[0].email == "john@example.com"
+    assert result_v2_full[0].bio == "Software developer"
+    assert result_v2_full[0].age == 28
+
+    # Test v2 with only required fields
+    parser_v2_minimal = PydanticToolsParser(tools=[UserV2])
+    message_v2_minimal = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_user_minimal",
+                "name": "UserV2",
+                "args": {"username": "jane_smith", "email": "jane@example.com"},
+            }
+        ],
+    )
+    generation_v2_minimal = ChatGeneration(message=message_v2_minimal)
+    result_v2_minimal = parser_v2_minimal.parse_result([generation_v2_minimal])
+
+    assert len(result_v2_minimal) == 1
+    assert isinstance(result_v2_minimal[0], UserV2)
+    assert result_v2_minimal[0].username == "jane_smith"
+    assert result_v2_minimal[0].email == "jane@example.com"
+    assert result_v2_minimal[0].bio is None
+    assert result_v2_minimal[0].age is None
+
+    # Test mixed v1 and v2 with partial optional fields
+    parser_mixed = PydanticToolsParser(tools=[ProductV1, UserV2])
+    message_mixed = AIMessage(
+        content="",
+        tool_calls=[
+            {
+                "id": "call_product",
+                "name": "ProductV1",
+                "args": {"name": "Keyboard", "price": 79.99, "stock": 100},
+            },
+            {
+                "id": "call_user",
+                "name": "UserV2",
+                "args": {
+                    "username": "alice",
+                    "email": "alice@example.com",
+                    "age": 35,
+                },
+            },
+        ],
+    )
+    generation_mixed = ChatGeneration(message=message_mixed)
+    result_mixed = parser_mixed.parse_result([generation_mixed])
+
+    assert len(result_mixed) == 2
+    assert isinstance(result_mixed[0], ProductV1)
+    assert result_mixed[0].name == "Keyboard"  # type: ignore[attr-defined,unused-ignore]
+    assert result_mixed[0].description is None  # type: ignore[attr-defined,unused-ignore]
+    assert result_mixed[0].stock == 100  # type: ignore[attr-defined,unused-ignore]
+    assert isinstance(result_mixed[1], UserV2)
+    assert result_mixed[1].username == "alice"
+    assert result_mixed[1].bio is None
+    assert result_mixed[1].age == 35

libs/core/tests/unit_tests/prompts/__snapshots__/test_chat.ambr

@@ -682,7 +682,7 @@
           
           May also hold extra provider-specific keys.
           
-          !!! version-added "Added in version 0.3.9"
+          !!! version-added "Added in `langchain-core` 0.3.9"
         ''',
         'properties': dict({
           'audio': dict({
@@ -800,7 +800,7 @@
           
           May also hold extra provider-specific keys.
           
-          !!! version-added "Added in version 0.3.9"
+          !!! version-added "Added in `langchain-core` 0.3.9"
         ''',
         'properties': dict({
           'audio': dict({
@@ -1319,10 +1319,12 @@
               }
               ```
           
-          !!! warning "Behavior changed in 0.3.9"
+          !!! warning "Behavior changed in `langchain-core` 0.3.9"
+          
               Added `input_token_details` and `output_token_details`.
           
           !!! note "LangSmith SDK"
+          
               The LangSmith SDK also has a `UsageMetadata` class. While the two share fields,
               LangSmith's `UsageMetadata` has additional fields to capture cost information
               used by the LangSmith platform.
@@ -2096,7 +2098,7 @@
           
           May also hold extra provider-specific keys.
           
-          !!! version-added "Added in version 0.3.9"
+          !!! version-added "Added in `langchain-core` 0.3.9"
         ''',
         'properties': dict({
           'audio': dict({
@@ -2214,7 +2216,7 @@
           
           May also hold extra provider-specific keys.
           
-          !!! version-added "Added in version 0.3.9"
+          !!! version-added "Added in `langchain-core` 0.3.9"
         ''',
         'properties': dict({
           'audio': dict({
@@ -2733,10 +2735,12 @@
               }
               ```
           
-          !!! warning "Behavior changed in 0.3.9"
+          !!! warning "Behavior changed in `langchain-core` 0.3.9"
+          
               Added `input_token_details` and `output_token_details`.
           
           !!! note "LangSmith SDK"
+          
               The LangSmith SDK also has a `UsageMetadata` class. While the two share fields,
               LangSmith's `UsageMetadata` has additional fields to capture cost information
               used by the LangSmith platform.

libs/core/tests/unit_tests/prompts/test_chat.py

@@ -1193,3 +1193,511 @@ def test_dict_message_prompt_template_errors_on_jinja2() -> None:
         _ = ChatPromptTemplate.from_messages(
             [("human", [prompt])], template_format="jinja2"
         )
+
+
+def test_rendering_prompt_with_conditionals_no_empty_text_blocks() -> None:
+    manifest = {
+        "lc": 1,
+        "type": "constructor",
+        "id": ["langchain_core", "prompts", "chat", "ChatPromptTemplate"],
+        "kwargs": {
+            "messages": [
+                {
+                    "lc": 1,
+                    "type": "constructor",
+                    "id": [
+                        "langchain_core",
+                        "prompts",
+                        "chat",
+                        "SystemMessagePromptTemplate",
+                    ],
+                    "kwargs": {
+                        "prompt": {
+                            "lc": 1,
+                            "type": "constructor",
+                            "id": [
+                                "langchain_core",
+                                "prompts",
+                                "prompt",
+                                "PromptTemplate",
+                            ],
+                            "kwargs": {
+                                "input_variables": [],
+                                "template_format": "mustache",
+                                "template": "Always echo back whatever I send you.",
+                            },
+                        },
+                    },
+                },
+                {
+                    "lc": 1,
+                    "type": "constructor",
+                    "id": [
+                        "langchain_core",
+                        "prompts",
+                        "chat",
+                        "HumanMessagePromptTemplate",
+                    ],
+                    "kwargs": {
+                        "prompt": [
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": [],
+                                    "template_format": "mustache",
+                                    "template": "Here is the teacher's prompt:",
+                                    "additional_content_fields": {
+                                        "text": "Here is the teacher's prompt:",
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": ["promptDescription"],
+                                    "template_format": "mustache",
+                                    "template": '"{{promptDescription}}"\n',
+                                    "additional_content_fields": {
+                                        "text": '"{{promptDescription}}"\n',
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": [],
+                                    "template_format": "mustache",
+                                    "template": "Here is the expected answer or success criteria given by the teacher:",  # noqa: E501
+                                    "additional_content_fields": {
+                                        "text": "Here is the expected answer or success criteria given by the teacher:",  # noqa: E501
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": ["expectedResponse"],
+                                    "template_format": "mustache",
+                                    "template": '"{{expectedResponse}}"\n',
+                                    "additional_content_fields": {
+                                        "text": '"{{expectedResponse}}"\n',
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": [],
+                                    "template_format": "mustache",
+                                    "template": "Note: This may be just one example of many possible correct ways for the student to respond.\n",  # noqa: E501
+                                    "additional_content_fields": {
+                                        "text": "Note: This may be just one example of many possible correct ways for the student to respond.\n",  # noqa: E501
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": [],
+                                    "template_format": "mustache",
+                                    "template": "For your evaluation of the student's response:\n",  # noqa: E501
+                                    "additional_content_fields": {
+                                        "text": "For your evaluation of the student's response:\n",  # noqa: E501
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": [],
+                                    "template_format": "mustache",
+                                    "template": "Here is a transcript of the student's explanation:",  # noqa: E501
+                                    "additional_content_fields": {
+                                        "text": "Here is a transcript of the student's explanation:",  # noqa: E501
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": ["responseTranscript"],
+                                    "template_format": "mustache",
+                                    "template": '"{{responseTranscript}}"\n',
+                                    "additional_content_fields": {
+                                        "text": '"{{responseTranscript}}"\n',
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": ["readingFluencyAnalysis"],
+                                    "template_format": "mustache",
+                                    "template": "{{#readingFluencyAnalysis}} For this task, the student's reading pronunciation and fluency were important. Here is analysis of the student's oral response: \"{{readingFluencyAnalysis}}\" {{/readingFluencyAnalysis}}",  # noqa: E501
+                                    "additional_content_fields": {
+                                        "text": "{{#readingFluencyAnalysis}} For this task, the student's reading pronunciation and fluency were important. Here is analysis of the student's oral response: \"{{readingFluencyAnalysis}}\" {{/readingFluencyAnalysis}}",  # noqa: E501
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": ["readingFluencyAnalysis"],
+                                    "template_format": "mustache",
+                                    "template": "{{#readingFluencyAnalysis}}Root analysis of the student's response (step 3) in this oral analysis rather than inconsistencies in the transcript.{{/readingFluencyAnalysis}}",  # noqa: E501
+                                    "additional_content_fields": {
+                                        "text": "{{#readingFluencyAnalysis}}Root analysis of the student's response (step 3) in this oral analysis rather than inconsistencies in the transcript.{{/readingFluencyAnalysis}}",  # noqa: E501
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": ["readingFluencyAnalysis"],
+                                    "template_format": "mustache",
+                                    "template": "{{#readingFluencyAnalysis}}Remember this is a student, so we care about general fluency - not voice acting. {{/readingFluencyAnalysis}}\n",  # noqa: E501
+                                    "additional_content_fields": {
+                                        "text": "{{#readingFluencyAnalysis}}Remember this is a student, so we care about general fluency - not voice acting. {{/readingFluencyAnalysis}}\n",  # noqa: E501
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": ["multipleChoiceAnalysis"],
+                                    "template_format": "mustache",
+                                    "template": "{{#multipleChoiceAnalysis}}Here is an analysis of the student's multiple choice response: {{multipleChoiceAnalysis}}{{/multipleChoiceAnalysis}}\n",  # noqa: E501
+                                    "additional_content_fields": {
+                                        "text": "{{#multipleChoiceAnalysis}}Here is an analysis of the student's multiple choice response: {{multipleChoiceAnalysis}}{{/multipleChoiceAnalysis}}\n",  # noqa: E501
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "prompt",
+                                    "PromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "input_variables": [],
+                                    "template_format": "mustache",
+                                    "template": "Here is the student's whiteboard:\n",
+                                    "additional_content_fields": {
+                                        "text": "Here is the student's whiteboard:\n",
+                                    },
+                                },
+                            },
+                            {
+                                "lc": 1,
+                                "type": "constructor",
+                                "id": [
+                                    "langchain_core",
+                                    "prompts",
+                                    "image",
+                                    "ImagePromptTemplate",
+                                ],
+                                "kwargs": {
+                                    "template": {
+                                        "url": "{{whiteboard}}",
+                                    },
+                                    "input_variables": ["whiteboard"],
+                                    "template_format": "mustache",
+                                    "additional_content_fields": {
+                                        "image_url": {
+                                            "url": "{{whiteboard}}",
+                                        },
+                                    },
+                                },
+                            },
+                        ],
+                        "additional_options": {},
+                    },
+                },
+            ],
+            "input_variables": [
+                "promptDescription",
+                "expectedResponse",
+                "responseTranscript",
+                "readingFluencyAnalysis",
+                "readingFluencyAnalysis",
+                "readingFluencyAnalysis",
+                "multipleChoiceAnalysis",
+                "whiteboard",
+            ],
+            "template_format": "mustache",
+            "metadata": {
+                "lc_hub_owner": "jacob",
+                "lc_hub_repo": "mustache-conditionals",
+                "lc_hub_commit_hash": "836ad82d512409ea6024fb760b76a27ba58fc68b1179656c0ba2789778686d46",  # noqa: E501
+            },
+        },
+    }
+
+    # Load the ChatPromptTemplate from the manifest
+    template = load(manifest)
+
+    # Format with conditional data - rules is empty, so mustache conditionals
+    # should not render
+    result = template.invoke(
+        {
+            "promptDescription": "What is the capital of the USA?",
+            "expectedResponse": "Washington, D.C.",
+            "responseTranscript": "Washington, D.C.",
+            "readingFluencyAnalysis": None,
+            "multipleChoiceAnalysis": "testing2",
+            "whiteboard": "https://foo.com/bar.png",
+        }
+    )
+    content = result.messages[1].content
+    assert isinstance(content, list)
+    assert not [
+        block for block in content if block["type"] == "text" and block["text"] == ""
+    ]
+
+
+def test_fstring_rejects_invalid_identifier_variable_names() -> None:
+    """Test that f-string templates block attribute access, indexing.
+
+    This validation prevents template injection attacks by blocking:
+    - Attribute access like {msg.__class__}
+    - Indexing like {msg[0]}
+    - All-digit variable names like {0} or {100} (interpreted as positional args)
+
+    While allowing any other field names that Python's Formatter accepts.
+    """
+    # Test that attribute access and indexing are blocked (security issue)
+    invalid_templates = [
+        "{msg.__class__}",  # Attribute access with dunder
+        "{msg.__class__.__name__}",  # Multiple dunders
+        "{msg.content}",  # Attribute access
+        "{msg[0]}",  # Item access
+        "{0}",  # All-digit variable name (positional argument)
+        "{100}",  # All-digit variable name (positional argument)
+        "{42}",  # All-digit variable name (positional argument)
+    ]
+
+    for template_str in invalid_templates:
+        with pytest.raises(ValueError, match="Invalid variable name") as exc_info:
+            ChatPromptTemplate.from_messages(
+                [("human", template_str)],
+                template_format="f-string",
+            )
+
+        error_msg = str(exc_info.value)
+        assert "Invalid variable name" in error_msg
+        # Check for any of the expected error message parts
+        assert (
+            "attribute access" in error_msg
+            or "indexing" in error_msg
+            or "positional arguments" in error_msg
+        )
+
+    # Valid templates - Python's Formatter accepts non-identifier field names
+    valid_templates = [
+        (
+            "Hello {name} and {user_id}",
+            {"name": "Alice", "user_id": "123"},
+            "Hello Alice and 123",
+        ),
+        ("User: {user-name}", {"user-name": "Bob"}, "User: Bob"),  # Hyphen allowed
+        (
+            "Value: {2fast}",
+            {"2fast": "Charlie"},
+            "Value: Charlie",
+        ),  # Starts with digit allowed
+        ("Data: {my var}", {"my var": "Dave"}, "Data: Dave"),  # Space allowed
+    ]
+
+    for template_str, kwargs, expected in valid_templates:
+        template = ChatPromptTemplate.from_messages(
+            [("human", template_str)],
+            template_format="f-string",
+        )
+        result = template.invoke(kwargs)
+        assert result.messages[0].content == expected  # type: ignore[attr-defined]
+
+
+def test_mustache_template_attribute_access_vulnerability() -> None:
+    """Test that Mustache template injection is blocked.
+
+    Verify the fix for security vulnerability GHSA-6qv9-48xg-fc7f
+
+    Previously, Mustache used getattr() as a fallback, allowing access to
+    dangerous attributes like __class__, __globals__, etc.
+
+    The fix adds isinstance checks that reject non-dict/list types.
+    When templates try to traverse Python objects, they get empty string
+    per Mustache spec (better than the previous behavior of exposing internals).
+    """
+    msg = HumanMessage("howdy")
+
+    # Template tries to access attributes on a Python object
+    prompt = ChatPromptTemplate.from_messages(
+        [("human", "{{question.__class__.__name__}}")],
+        template_format="mustache",
+    )
+
+    # After the fix: returns empty string (attack blocked!)
+    # Previously would return "HumanMessage" via getattr()
+    result = prompt.invoke({"question": msg})
+    assert result.messages[0].content == ""  # type: ignore[attr-defined]
+
+    # Mustache still works correctly with actual dicts
+    prompt_dict = ChatPromptTemplate.from_messages(
+        [("human", "{{person.name}}")],
+        template_format="mustache",
+    )
+    result_dict = prompt_dict.invoke({"person": {"name": "Alice"}})
+    assert result_dict.messages[0].content == "Alice"  # type: ignore[attr-defined]
+
+
+@pytest.mark.requires("jinja2")
+def test_jinja2_template_attribute_access_is_blocked() -> None:
+    """Test that Jinja2 SandboxedEnvironment blocks dangerous attribute access.
+
+    This test verifies that Jinja2's sandbox successfully blocks access to
+    dangerous dunder attributes like __class__, unlike Mustache.
+
+    GOOD: Jinja2 SandboxedEnvironment raises SecurityError when attempting
+    to access __class__, __globals__, etc. This is expected behavior.
+    """
+    msg = HumanMessage("howdy")
+
+    # Create a Jinja2 template that attempts to access __class__.__name__
+    prompt = ChatPromptTemplate.from_messages(
+        [("human", "{{question.__class__.__name__}}")],
+        template_format="jinja2",
+    )
+
+    # Jinja2 sandbox should block this with SecurityError
+    with pytest.raises(Exception, match="attribute") as exc_info:
+        prompt.invoke(
+            {"question": msg, "question.__class__.__name__": "safe_placeholder"}
+        )
+
+    # Verify it's a SecurityError from Jinja2 blocking __class__ access
+    error_msg = str(exc_info.value)
+    assert (
+        "SecurityError" in str(type(exc_info.value))
+        or "access to attribute '__class__'" in error_msg
+    ), f"Expected SecurityError blocking __class__, got: {error_msg}"
+
+
+@pytest.mark.requires("jinja2")
+def test_jinja2_blocks_all_attribute_access() -> None:
+    """Test that Jinja2 now blocks ALL attribute/method access for security.
+
+    After the fix, Jinja2 uses _RestrictedSandboxedEnvironment which blocks
+    ALL attribute access, not just dunder attributes. This prevents the
+    parse_raw() vulnerability.
+    """
+    msg = HumanMessage("test content")
+
+    # Test 1: Simple variable access should still work
+    prompt_simple = ChatPromptTemplate.from_messages(
+        [("human", "Message: {{message}}")],
+        template_format="jinja2",
+    )
+    result = prompt_simple.invoke({"message": "hello world"})
+    assert "hello world" in result.messages[0].content  # type: ignore[attr-defined]
+
+    # Test 2: Attribute access should now be blocked (including safe attributes)
+    prompt_attr = ChatPromptTemplate.from_messages(
+        [("human", "Content: {{msg.content}}")],
+        template_format="jinja2",
+    )
+    with pytest.raises(Exception, match="attribute") as exc_info:
+        prompt_attr.invoke({"msg": msg})
+
+    error_msg = str(exc_info.value)
+    assert (
+        "SecurityError" in str(type(exc_info.value))
+        or "Access to attributes is not allowed" in error_msg
+    ), f"Expected SecurityError blocking attribute access, got: {error_msg}"

libs/core/tests/unit_tests/prompts/test_structured.py

@@ -125,7 +125,9 @@ def test_structured_prompt_kwargs() -> None:
 
 def test_structured_prompt_template_format() -> None:
     prompt = StructuredPrompt(
-        [("human", "hi {{person.name}}")], schema={}, template_format="mustache"
+        [("human", "hi {{person.name}}")],
+        schema={"type": "object", "properties": {}, "title": "foo"},
+        template_format="mustache",
     )
     assert prompt.messages[0].prompt.template_format == "mustache"  # type: ignore[union-attr, union-attr]
     assert prompt.input_variables == ["person"]
@@ -136,4 +138,8 @@ def test_structured_prompt_template_format() -> None:
 
 def test_structured_prompt_template_empty_vars() -> None:
     with pytest.raises(ChevronError, match="empty tag"):
-        StructuredPrompt([("human", "hi {{}}")], schema={}, template_format="mustache")
+        StructuredPrompt(
+            [("human", "hi {{}}")],
+            schema={"type": "object", "properties": {}, "title": "foo"},
+            template_format="mustache",
+        )