Add short flag -m for --metrics-port to improve discoverability.
Add K8SGPT_METRICS_PORT environment variable support, consistent
with other K8SGPT_* environment variables.
This helps users who encounter port conflicts on the default
metrics port (8081) when running k8sgpt serve with --mcp or
other configurations.
Signed-off-by: Three Foxes (in a Trenchcoat) <threefoxes53235@gmail.com>
Signed-off-by: Three Foxes (in a Trenchcoat) <threefoxesyes3inatrenchcoat@gmail.com>
* fix: update OpenAI API key generation URL to reflect new platform link
Updated the outdated URL 'https://beta.openai.com/account/api-keys' to the current OpenAI API key generation page 'https://platform.openai.com/account/api-keys'.
This resolves the issue where users were directed to an incorrect URL when generating an OpenAI API key.
Signed-off-by: 100daysofdevops <47483190+100daysofdevops@users.noreply.github.com>
* fix(deps):Add transition plan for GPT-3.5 Turbo to GPT-4o
- A comprehensive comparison of GPT-3.5 Turbo and GPT-4o models, focusing on performance and cost improvements.
- Documentation updates highlighting the planned deprecation of gpt-3.5-turbo-0301 on February 13, 2025.
- Clear migration guidelines for transitioning to GPT-4o or GPT-4o mini to ensure service continuity.
Signed-off-by: 100daysofdevops <47483190+100daysofdevops@users.noreply.github.com>
---------
Signed-off-by: 100daysofdevops <47483190+100daysofdevops@users.noreply.github.com>
Co-authored-by: Alex Jones <alexsimonjones@gmail.com>
* feat: add stats option to analyze command for performance insights
Introduced a new feature to the analyze command that enables users to print detailed performance statistics of each analyzer. This enhancement aids in debugging and understanding the time taken by various components during analysis, providing valuable insights for performance optimization.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: enhance analysis command with statistics option
Refactored the analysis command to support an enhanced statistics option, enabling users to opt-in for detailed performance metrics of the analysis process. This change introduces a more flexible approach to handling statistics, allowing for a clearer separation between the analysis output and performance metrics, thereby improving the usability and insights provided to the user.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
---------
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
Co-authored-by: Alex Jones <alexsimonjones@gmail.com>
Co-authored-by: Aris Boutselis <arisboutselis08@gmail.com>
* feat: add custom analyzer management capability
Introduced the ability to manage custom analyzers in the K8sGPT application, enabling users to add, deploy, and configure custom analyzers from various sources. This enhancement supports extending the application's analytical capabilities by integrating external analysis tools, thus offering more flexibility and customization options to meet specific user needs.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: enhance custom analyzer management with removal functionality
Introduced the ability to remove custom analyzers, streamlining the management process and ensuring flexibility in custom analyzer configuration. This enhancement addresses the need for dynamic customization and maintenance of analyzer setups, facilitating easier updates and modifications to the analysis environment.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: add list command to customAnalyzer for displaying configured analyzers
Implemented a new list command within the customAnalyzer module to enable users to view all configured custom analyzers. This enhancement aims to improve usability by providing a straightforward method for users to inspect their custom analyzer configurations directly from the command line.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: add support for listing, adding, and removing custom analyzers
This update introduces commands to manage custom analyzers in the k8sgpt tool, enhancing flexibility and control over analyzer configurations without the need for direct installation or docker dependency.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: support private docker image authentication for custom analyzers
Added authentication support for pulling private Docker images when adding custom analyzers, enhancing security and access control.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: remove Docker custom analyzer installation
Removed the installation and deployment functionality for custom analyzers, streamlining the process of adding analyzers. This change focuses on simplifying the configuration by eliminating the need for specifying installation types, package URLs, and authentication details for Docker images. The goal is to enhance user experience by making the addition of custom analyzers more straightforward and less error-prone.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* fix: remove unused packageUrl
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: update add command description to reflect broader functionality
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: Add name validation for custom analyzer creation
To ensure the integrity and consistency of analyzer names, we introduced a validation step that checks the format of the name against a predefined regex pattern. This change aims to prevent the creation of analyzers with invalid names, enhancing the system's reliability and usability.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
* feat: refactor customAnalyzer package for consistent naming
Refactored the customAnalyzer package and its references to use consistent snake_case naming for improved code readability and alignment with Go naming conventions.
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
---------
Signed-off-by: Matthis Holleville <matthish29@gmail.com>
This commit adds Top-K sampling, a feature that allows users to control
the randomness of the generated text by specifying the number of most
probable next words considered by the model. This enhances user control
and potentially improves the quality of the generated outputs.
Fixes: https://github.com/k8sgpt-ai/k8sgpt/issues/1105
Signed-off-by: VaibhavMalik4187 <vaibhavmalik2018@gmail.com>
Co-authored-by: Alex Jones <alexsimonjones@gmail.com>
* chore: allows an environmental override of the default AWS region and using it for bedrock
Signed-off-by: Alex Jones <alexsimonjones@gmail.com>
* chore: missing provider region
Signed-off-by: Alex Jones <alexsimonjones@gmail.com>
---------
Signed-off-by: Alex Jones <alexsimonjones@gmail.com>
Now, the default value of the `backend` flag for the analyze command
will be an empty string. And the `NewAnalysis` function has been
modified to use the default backend set by the user if the backend flag
is not provided and the `defaultprovider` is set in the config file.
Otherwise, backend will be set to "openai".
Fixes: https://github.com/k8sgpt-ai/k8sgpt/issues/902
Signed-off-by: VaibhavMalik4187 <vaibhavmalik2018@gmail.com>
Co-authored-by: JuHyung Son <sonju0427@gmail.com>
Removed the shorthand for the `http` flag in the serve command because
it was contradicting with the shorthand of the `help` command which is
automatically added on execution if the `help` flag is not already
defined.
Fixes: https://github.com/k8sgpt-ai/k8sgpt/issues/968
Signed-off-by: VaibhavMalik4187 <vaibhavmalik2018@gmail.com>
* chore: linting improvements and catching false positives
Signed-off-by: Alex Jones <alexsimonjones@gmail.com>
* chore: linting improvements and catching false positives
Signed-off-by: Alex Jones <alexsimonjones@gmail.com>
* chore: linting improvements and catching false positives
Signed-off-by: Alex Jones <alexsimonjones@gmail.com>
* chore: increase linter time out
Signed-off-by: Alex Jones <alexsimonjones@gmail.com>
---------
Signed-off-by: Alex Jones <alexsimonjones@gmail.com>
