Repolode
The-PR-Agent

The-PR-Agent/pr-agent

Open-source AI agent that automatically reviews pull requests and offers /describe, /review, /improve and /ask commands on GitHub, GitLab, Bitbucket and Azure DevOps.

Open-source alternative to:CodeRabbit

Stars
13,463
Forks
1,972
Open issues
28
Last commit
5 hours ago

What is it?

PR-Agent is an open-source tool that uses large language models to review pull requests. Each tool, such as /review, /improve, /describe and /ask, runs as a comment on a PR or from the command line and makes a single LLM call per invocation, so it is designed to be quick and inexpensive to run.

It works with GitHub, GitLab, Bitbucket, Azure DevOps and Gitea, and can be deployed as a GitHub Action, a CLI tool, a Docker container, or a self-hosted webhook service. Model access goes through LiteLLM, which covers OpenAI, Anthropic Claude, Google Gemini, DeepSeek, Mistral, Ollama and other providers. Prompts and review categories are configured through TOML files, so teams can adjust review behavior to their needs.

The project is community-maintained and was donated by Qodo, which now offers its own hosted AI review product separately.

Key features

  • 01Provides /describe, /review, /improve and /ask commands for pull requests.
  • 02Supports GitHub, GitLab, Bitbucket, Azure DevOps and Gitea.
  • 03Connects to OpenAI, Claude, Gemini, Ollama and other models through LiteLLM.
  • 04Can run as a GitHub Action, CLI, Docker container or self-hosted webhook.
  • 05Customizes prompts and review categories through configuration files.

Use cases

  • Adding automated first-pass reviews to every pull request in a GitHub repository.
  • Running code review locally on a PR before requesting human review.
  • Self-hosting an AI reviewer to keep code and data inside your own infrastructure.

Quick start

  1. 1

    Add the PR Agent GitHub Action workflow file to your repository

  2. 2

    Install the CLI with pip including the GitHub extra

    pip install "pr-agent[github]"
  3. 3

    Set your OpenAI key in the environment

    export OPENAI_KEY=your_key_here
  4. 4

    Run a review on a pull request from the CLI

    pr-agent --pr_url https://github.com/owner/repo/pull/123 review

Things to know

  • Requires an API key from an LLM provider such as OpenAI, which you pay for.
  • Git provider SDKs are optional extras, so you must install the one matching your platform.
  • The /help_docs command is temporarily disabled due to a credential-exposure issue.

Topics

FAQ

What is pr-agent?
Open-source AI agent that automatically reviews pull requests and offers /describe, /review, /improve and /ask commands on GitHub, GitLab, Bitbucket and Azure DevOps. PR-Agent is an open-source tool that uses large language models to review pull requests. Each tool, such as /review, /improve, /describe and /ask, runs as a comment on a PR or from the command line and makes a single LLM call per invocation, so it is designed to be quick and inexpensive to run.
Is pr-agent free to use? What license does it have?
pr-agent is open source under the MIT license. Check the license terms before using it in commercial products.
Is pr-agent still maintained?
Actively maintained. The last commit to pr-agent was on Oct 20, 2026.
What can I use pr-agent for?
Adding automated first-pass reviews to every pull request in a GitHub repository. Running code review locally on a PR before requesting human review. Self-hosting an AI reviewer to keep code and data inside your own infrastructure.
How do I install pr-agent?
Add the PR Agent GitHub Action workflow file to your repository Install the CLI with pip including the GitHub extra: pip install "pr-agent[github]" Set your OpenAI key in the environment: export OPENAI_KEY=your_key_here Run a review on a pull request from the CLI: pr-agent --pr_url https://github.com/owner/repo/pull/123 review
Where can I get pr-agent?
The source code and installation instructions are on GitHub: https://github.com/The-PR-Agent/pr-agent

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects