Debuly

OpenRabbit

Free self-hosted GitHub pull request reviewer that runs in your own Actions and posts AI summaries and inline comments

Open source

What is OpenRabbit?

OpenRabbit is a free, open-source GitHub pull request reviewer that runs inside your own GitHub Actions environment, so there is no server to manage and no subscription. When a pull request is opened or updated, it reads the diff, asks a pluggable LLM provider such as OpenRouter or Groq to analyze it, and posts a structured review with a readable summary and inline comments or suggestions. To try to avoid judging the diff in isolation, it can fetch extra files in a second pass and read linked GitHub issues to understand what the change should do. It also flags unrelated edits that look like drift and checks suggestions against your code before showing them. Setup is one workflow file plus an LLM_API_KEY stored as a GitHub Actions secret. The default OpenRouter free model pool costs nothing but is subject to rate limits, so your own key is recommended. The project is MIT licensed.

Use cases

  • Review GitHub pull requests automatically
  • Get inline suggestions on code diffs
  • Audit pull requests for security issues
  • Spot performance problems before merging
  • Keep unrelated changes out of a pull request
  • Replace a paid AI review tool with a free one

Who it's for

  • Open-source maintainers
  • Small dev teams
  • Solo developers

Why use OpenRabbit

  • No subscription and no server to host, since the review runs in your own CI.
  • Code stays in your runner and is not proxied or stored by a central service.
  • Choose your own LLM provider and model rather than being tied to one vendor.
  • Context from linked issues and extra files makes feedback more relevant than diff-only review.
  • Setup takes a single workflow file and one repository secret.

OpenRabbit features

  • Runs entirely in your own GitHub Actions runner
  • Pluggable LLM providers, including OpenRouter and Groq
  • Summary, inline, or both review modes
  • Two-stage file fetch to pull in missing context
  • Linked GitHub issue awareness
  • Socratic mentor mode that asks guiding questions
  • Security and performance review focus
  • Suggestion validation against your code
  • DRIFT detection for unrelated changes
  • Configurable tone setting

OpenRabbit integrations

  • GitHub
  • OpenRouter
  • Groq
  • GitHub Actions

How to get started with OpenRabbit

  1. Create .github/workflows/reviewer.yml and paste the OpenRabbit workflow.
  2. Add your LLM API key as a repository secret named LLM_API_KEY.
  3. Set the provider, model, review mode and tone in the workflow inputs.
  4. When a pull request opens or updates, the action reads the diff and fetches extra context if needed.
  5. OpenRabbit posts a summary and inline comments or suggestions on the pull request.

OpenRabbit limitations

  • The default free OpenRouter model pool is subject to rate limits and can be inconsistent
  • Works only with GitHub pull requests
  • You provide and pay for your own LLM API key if you want more reliable results

OpenRabbit pricing

OpenRabbit is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

OpenRabbit alternatives

The main OpenRabbit alternatives listed on Debuly are CodeRabbit, GitHub Copilot, PR-Agent, Greptile, Sourcery and Gito. The table compares pricing and platforms; order is A–Z.

OpenRabbit compared with its alternatives
ToolPricingStarting pricePlatforms
OpenRabbitOpen source—GitHub Actions, Self-hosted
CodeRabbitFreemium$24/moWeb, VS Code extension, CLI, Self-hosted
GitHub CopilotFreemium$10/mo—
GitoOpen source—CLI, Windows, GitHub Actions, GitLab CI
GreptileFreemium$1, usage-basedWeb, API, CLI, Self-hosted
PR-AgentOpen source—CLI, Docker, Self-hosted, GitHub
SourceryFreemium$12/moWeb, GitHub, GitLab, API
  1. CodeRabbit

    AI code review for pull requests, IDEs and the CLI, with pre-merge checks, one-click fixes, triage and security scans.

  2. GitHub Copilot

    GitHub's AI coding assistant for code completion, chat and agents in VS Code, JetBrains, Visual Studio, the CLI and GitHub.com.

  3. Gito

    Open-source AI code reviewer that checks GitHub pull requests and local changes with the language model provider you choose.

  4. Greptile

    AI pull request reviewer that maps your whole repository, so each review weighs the change against code outside the diff.

  5. PR-Agent

    Open-source AI agent that describes, reviews and suggests improvements on pull requests across GitHub, GitLab and more.

  6. Sourcery

    AI code review for GitHub and GitLab pull requests with summaries, suggested fixes and security scanning, free on public repositories.

OpenRabbit FAQ

How much does OpenRabbit cost?

OpenRabbit itself is free and MIT licensed. You can use the OpenRouter free model pool at no charge, or pay your own LLM provider for a key.

Is there a free plan for OpenRabbit?

There are no plans or tiers. The tool is free, and the default setup uses a free OpenRouter model, though it can hit rate limits.

Which LLM providers does OpenRabbit support?

It supports pluggable providers, with Groq and OpenRouter named in the setup. New provider adapters can be added under src/llm.

Where does my code go when OpenRabbit reviews a pull request?

It runs in your own GitHub Actions runner. The project says code is not proxied or stored by a central authority, though the diff is sent to the LLM provider you choose.

Is OpenRabbit open source?

Yes. It is licensed under MIT and the source is on GitHub, where you can open issues or pull requests.

What are alternatives to OpenRabbit?

Alternatives include CodeRabbit, GitHub Copilot code review, PR-Agent, Greptile and Sourcery.

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

SourceTypeAccessed
github.com/aryanbrite/openrabbitofficial

Image credits

  • Logo: none collected yet, so a monogram is shown.
  • Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.