{ "cells": [ { "cell_type": "markdown", "id": "517a9fd4", "metadata": {}, "source": [ "# BabyAGI User Guide\n", "\n", "This notebook demonstrates how to implement [BabyAGI](https://github.com/yoheinakajima/babyagi/tree/main) by [Yohei Nakajima](https://twitter.com/yoheinakajima). BabyAGI is an AI agent that can generate and pretend to execute tasks based on a given objective.\n", "\n", "This guide will help you understand the components to create your own recursive agents.\n" ] }, { "cell_type": "markdown", "id": "556af556", "metadata": {}, "source": [ "## Install and Import Required Modules" ] }, { "cell_type": "code", "execution_count": 1, "id": "33a0c80a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Note: you may need to restart the kernel to use updated packages.\n" ] } ], "source": [ "%pip install pinecone-client > /dev/null" ] }, { "cell_type": "code", "execution_count": 2, "id": "c8a354b6", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/Users/wfh/code/lc/lckg/.venv/lib/python3.11/site-packages/pinecone/index.py:4: TqdmExperimentalWarning: Using `tqdm.autonotebook.tqdm` in notebook mode. Use `tqdm.tqdm` instead to force console mode (e.g. in jupyter console)\n", " from tqdm.autonotebook import tqdm\n" ] } ], "source": [ "import os\n", "from collections import deque\n", "from typing import Dict, List, Optional\n", "\n", "import pinecone\n", "from langchain import LLMChain, OpenAI, PromptTemplate\n", "from langchain.embeddings import OpenAIEmbeddings\n", "from langchain.llms import BaseLLM\n", "from langchain.vectorstores import Pinecone\n", "from langchain.vectorstores.base import VectorStore\n", "from pydantic import BaseModel, Field\n" ] }, { "cell_type": "markdown", "id": "09f70772", "metadata": {}, "source": [ "## Connect to the Vector Index\n", "\n", "Define the required environment, and connect to the vector index.\n", "\n", "See [Pinecone](https://app.pinecone.io/) to create your API key." ] }, { "cell_type": "code", "execution_count": 3, "id": "794045d4", "metadata": {}, "outputs": [], "source": [ "# Define the parameters for this run.\n", "index_name = \"langchain-baby-agi\"\n", "OBJECTIVE = \"Interstellar Travel\" # Ultimate goal\n", "YOUR_FIRST_TASK = \"brush my teeth\" # Where to start" ] }, { "cell_type": "code", "execution_count": 4, "id": "802b797e", "metadata": {}, "outputs": [], "source": [ "llm = OpenAI() # The underlying LLM\n", "\n", "# Get Pinecone environment\n", "pinecone.init(\n", " api_key=os.environ[\"PINECONE_API_KEY\"], # find at app.pinecone.io\n", " environment=os.environ[\"PINECONE_ENVIRONMENT\"] # next to api key in console\n", ")\n", "\n", "# Define your embedding model\n", "embeddings_model = OpenAIEmbeddings()\n", "\n", "# Create Pinecone index if it doesn't exist\n", "dimension = 1536 # dimension of the embedding space\n", "metric = \"cosine\" # distance function\n", "pod_type = \"p1\" # performance-optimized\n", "if index_name not in pinecone.list_indexes():\n", " pinecone.create_index(index_name, dimension=dimension, metric=metric, pod_type=pod_type)" ] }, { "cell_type": "code", "execution_count": 5, "id": "30e8ecc4", "metadata": { "scrolled": true }, "outputs": [], "source": [ "# Fetch the index\n", "text_key = \"page_content\"\n", "if index_name in pinecone.list_indexes():\n", " index = pinecone.Index(index_name)\n", "else:\n", " raise ValueError(f\"Index '{index_name}' not found in your Pinecone project.\")\n", "\n", "vectorstore = Pinecone(index, embeddings_model.embed_query, text_key)" ] }, { "cell_type": "markdown", "id": "0f3b72bf", "metadata": {}, "source": [ "## Define the Chains\n", "\n", "BabyAGI relies on three LLM chains:\n", "- Task creation chain to select new tasks to add to the list\n", "- Task prioritization chain to re-prioritize tasks\n", "- Execution Chain to execute the tasks" ] }, { "cell_type": "code", "execution_count": 6, "id": "bf4bd5cd", "metadata": {}, "outputs": [], "source": [ "class TaskCreationChain(LLMChain):\n", " \"\"\"Chain to generates tasks.\"\"\"\n", "\n", " @classmethod\n", " def from_llm(cls, llm: BaseLLM, objective: str, verbose: bool = True) -> LLMChain:\n", " \"\"\"Get the response parser.\"\"\"\n", " task_creation_template = (\n", " \"You are an task creation AI that uses the result of an execution agent\"\n", " \" to create new tasks with the following objective: {objective},\"\n", " \" The last completed task has the result: {result}.\"\n", " \" This result was based on this task description: {task_description}.\"\n", " \" These are incomplete tasks: {incomplete_tasks}.\"\n", " \" Based on the result, create new tasks to be completed\"\n", " \" by the AI system that do not overlap with incomplete tasks.\"\n", " \" Return the tasks as an array.\"\n", " )\n", " prompt = PromptTemplate(\n", " template=task_creation_template,\n", " partial_variables={\"objective\": objective},\n", " input_variables=[\"result\", \"task_description\", \"incomplete_tasks\"],\n", " )\n", " return cls(prompt=prompt, llm=llm, verbose=verbose)\n", " \n", " def get_next_task(self, result: Dict, task_description: str, task_list: List[str]) -> List[Dict]:\n", " \"\"\"Get the next task.\"\"\"\n", " incomplete_tasks = \", \".join(task_list)\n", " response = self.run(result=result, task_description=task_description, incomplete_tasks=incomplete_tasks)\n", " new_tasks = response.split('\\n')\n", " return [{\"task_name\": task_name} for task_name in new_tasks if task_name.strip()]" ] }, { "cell_type": "code", "execution_count": 7, "id": "b6488ffe", "metadata": {}, "outputs": [], "source": [ "class TaskPrioritizationChain(LLMChain):\n", " \"\"\"Chain to prioritize tasks.\"\"\"\n", "\n", " @classmethod\n", " def from_llm(cls, llm: BaseLLM, objective: str, verbose: bool = True) -> LLMChain:\n", " \"\"\"Get the response parser.\"\"\"\n", " task_prioritization_template = (\n", " \"You are an task prioritization AI tasked with cleaning the formatting of and reprioritizing\"\n", " \" the following tasks: {task_names}.\"\n", " \" Consider the ultimate objective of your team: {objective}.\"\n", " \" Do not remove any tasks. Return the result as a numbered list, like:\"\n", " \" #. First task\"\n", " \" #. Second task\"\n", " \" Start the task list with number {next_task_id}.\"\n", " )\n", " prompt = PromptTemplate(\n", " template=task_prioritization_template,\n", " partial_variables={\"objective\": objective},\n", " input_variables=[\"task_names\", \"next_task_id\"],\n", " )\n", " return cls(prompt=prompt, llm=llm, verbose=verbose)\n", "\n", " def prioritize_tasks(self, this_task_id: int, task_list: List[Dict]) -> List[Dict]:\n", " \"\"\"Prioritize tasks.\"\"\"\n", " task_names = [t[\"task_name\"] for t in task_list]\n", " next_task_id = int(this_task_id) + 1\n", " response = self.run(task_names=task_names, next_task_id=next_task_id)\n", " new_tasks = response.split('\\n')\n", " prioritized_task_list = []\n", " for task_string in new_tasks:\n", " if not task_string.strip():\n", " continue\n", " task_parts = task_string.strip().split(\".\", 1)\n", " if len(task_parts) == 2:\n", " task_id = task_parts[0].strip()\n", " task_name = task_parts[1].strip()\n", " prioritized_task_list.append({\"task_id\": task_id, \"task_name\": task_name})\n", " return prioritized_task_list\n" ] }, { "cell_type": "code", "execution_count": 8, "id": "b43cd580", "metadata": {}, "outputs": [], "source": [ "class ExecutionChain(LLMChain):\n", " \"\"\"Chain to execute tasks.\"\"\"\n", " \n", " vectorstore: VectorStore = Field(init=False)\n", "\n", " @classmethod\n", " def from_llm(cls, llm: BaseLLM, vectorstore: VectorStore, verbose: bool = True) -> LLMChain:\n", " \"\"\"Get the response parser.\"\"\"\n", " execution_template = (\n", " \"You are an AI who performs one task based on the following objective: {objective}.\"\n", " \" Take into account these previously completed tasks: {context}.\"\n", " \" Your task: {task}.\"\n", " \" Response:\"\n", " )\n", " prompt = PromptTemplate(\n", " template=execution_template,\n", " input_variables=[\"objective\", \"context\", \"task\"],\n", " )\n", " return cls(prompt=prompt, llm=llm, verbose=verbose, vectorstore=vectorstore)\n", " \n", " def _get_top_tasks(self, query: str, k: int) -> List[str]:\n", " \"\"\"Get the top k tasks based on the query.\"\"\"\n", " results = self.vectorstore.similarity_search_with_score(query, k=k)\n", " if not results:\n", " return []\n", " sorted_results, _ = zip(*sorted(results, key=lambda x: x[1], reverse=True))\n", " return [str(item.metadata['task']) for item in sorted_results]\n", " \n", " def execute_task(self, objective: str, task: str, k: int = 5) -> str:\n", " \"\"\"Execute a task.\"\"\"\n", " context = self._get_top_tasks(query=objective, k=k)\n", " return self.run(objective=objective, context=context, task=task)\n" ] }, { "cell_type": "markdown", "id": "3ad996c5", "metadata": {}, "source": [ "### Define the BabyAGI Controller\n", "\n", "BabyAGI composes the chains defined above in a (potentially-)infinite loop." ] }, { "cell_type": "code", "execution_count": 9, "id": "1e978938", "metadata": {}, "outputs": [], "source": [ "class BabyAGI(BaseModel):\n", " \"\"\"Controller model for the BabyAGI agent.\"\"\"\n", "\n", " objective: str = Field(alias=\"objective\")\n", " task_list: deque = Field(default_factory=deque)\n", " task_creation_chain: TaskCreationChain = Field(...)\n", " task_prioritization_chain: TaskPrioritizationChain = Field(...)\n", " execution_chain: ExecutionChain = Field(...)\n", " task_id_counter: int = Field(1)\n", "\n", " def add_task(self, task: Dict):\n", " self.task_list.append(task)\n", "\n", " def print_task_list(self):\n", " print(\"\\033[95m\\033[1m\" + \"\\n*****TASK LIST*****\\n\" + \"\\033[0m\\033[0m\")\n", " for t in self.task_list:\n", " print(str(t[\"task_id\"]) + \": \" + t[\"task_name\"])\n", "\n", " def print_next_task(self, task: Dict):\n", " print(\"\\033[92m\\033[1m\" + \"\\n*****NEXT TASK*****\\n\" + \"\\033[0m\\033[0m\")\n", " print(str(task[\"task_id\"]) + \": \" + task[\"task_name\"])\n", "\n", " def print_task_result(self, result: str):\n", " print(\"\\033[93m\\033[1m\" + \"\\n*****TASK RESULT*****\\n\" + \"\\033[0m\\033[0m\")\n", " print(result)\n", "\n", " def run(self, max_iterations: Optional[int] = None):\n", " \"\"\"Run the agent.\"\"\"\n", " num_iters = 0\n", " while True:\n", " if self.task_list:\n", " self.print_task_list()\n", "\n", " # Step 1: Pull the first task\n", " task = self.task_list.popleft()\n", " self.print_next_task(task)\n", "\n", " # Step 2: Execute the task\n", " result = self.execution_chain.execute_task(\n", " self.objective, task[\"task_name\"]\n", " )\n", " this_task_id = int(task[\"task_id\"])\n", " self.print_task_result(result)\n", "\n", " # Step 3: Store the result in Pinecone\n", " result_id = f\"result_{task['task_id']}\"\n", " self.execution_chain.vectorstore.add_texts(\n", " texts=[result],\n", " metadatas=[{\"task\": task[\"task_name\"]}],\n", " ids=[result_id],\n", " )\n", "\n", " # Step 4: Create new tasks and reprioritize task list\n", " new_tasks = self.task_creation_chain.get_next_task(\n", " result, task[\"task_name\"], [t[\"task_name\"] for t in self.task_list]\n", " )\n", " for new_task in new_tasks:\n", " self.task_id_counter += 1\n", " new_task.update({\"task_id\": self.task_id_counter})\n", " self.add_task(new_task)\n", " self.task_list = deque(\n", " self.task_prioritization_chain.prioritize_tasks(\n", " this_task_id, list(self.task_list)\n", " )\n", " )\n", " num_iters += 1\n", " if max_iterations is not None and num_iters == max_iterations:\n", " print(\"\\033[91m\\033[1m\" + \"\\n*****TASK ENDING*****\\n\" + \"\\033[0m\\033[0m\")\n", " break\n", "\n", " @classmethod\n", " def from_llm_and_objectives(\n", " cls,\n", " llm: BaseLLM,\n", " vectorstore: VectorStore,\n", " objective: str,\n", " first_task: str,\n", " verbose: bool = False,\n", " ) -> \"BabyAGI\":\n", " \"\"\"Initialize the BabyAGI Controller.\"\"\"\n", " task_creation_chain = TaskCreationChain.from_llm(\n", " llm, objective, verbose=verbose\n", " )\n", " task_prioritization_chain = TaskPrioritizationChain.from_llm(\n", " llm, objective, verbose=verbose\n", " )\n", " execution_chain = ExecutionChain.from_llm(llm, vectorstore, verbose=verbose)\n", " controller = cls(\n", " objective=objective,\n", " task_creation_chain=task_creation_chain,\n", " task_prioritization_chain=task_prioritization_chain,\n", " execution_chain=execution_chain,\n", " )\n", " controller.add_task({\"task_id\": 1, \"task_name\": first_task})\n", " return controller" ] }, { "cell_type": "markdown", "id": "05ba762e", "metadata": {}, "source": [ "### Run the BabyAGI\n", "\n", "Now it's time to create the BabyAGI controller and watch it try to accomplish your objective." ] }, { "cell_type": "code", "execution_count": 10, "id": "3d69899b", "metadata": {}, "outputs": [], "source": [ "# Run the agent\n", "verbose=False\n", "\n", "baby_agi = BabyAGI.from_llm_and_objectives(\n", " llm=llm,\n", " vectorstore=vectorstore,\n", " objective=OBJECTIVE,\n", " first_task=YOUR_FIRST_TASK,\n", " verbose=verbose\n", ")" ] }, { "cell_type": "code", "execution_count": 11, "id": "f7957b51", "metadata": { "scrolled": false }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\u001b[95m\u001b[1m\n", "*****TASK LIST*****\n", "\u001b[0m\u001b[0m\n", "1: brush my teeth\n", "\u001b[92m\u001b[1m\n", "*****NEXT TASK*****\n", "\u001b[0m\u001b[0m\n", "1: brush my teeth\n", "\u001b[93m\u001b[1m\n", "*****TASK RESULT*****\n", "\u001b[0m\u001b[0m\n", " The objective of this task does not align with the objective of Interstellar Travel. Please provide a task that is related to Interstellar Travel.\n" ] }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "68b489b48aad4ce886088bd8be140bfa", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Upserted vectors: 0%| | 0/1 [00:00