Repolode
Submit
Nayjest

Nayjest/Gito

Open-source AI code reviewer that runs LLM-based reviews on GitHub pull requests and local branch changes, with any model provider.

Open-source alternative to:CodeRabbit

View on GitHubWebsite
Stars
466
Forks
45
Open issues
61
Last commit
last month

What is it?

Gito is a command-line AI code reviewer that analyzes pull requests or local branch changes with a language model and reports likely security issues, bugs, and maintainability problems. It is built on ai-microcore, so it works with OpenAI-compatible APIs, Anthropic, Google, and local runtimes such as Ollama, vLLM, and llama.cpp.

In CI, Gito can be set up through GitHub Actions or GitLab CI workflows, where it posts review results as PR comments and responds to maintainer mentions. The bundled `gito deploy` wizard generates the workflow files. Locally, `gito review` checks the current branch against the main branch, and `gito remote` reviews a remote repository.

Behavior is configured per repository through `.gito/config.toml`, which can customize prompts, output templates, file exclusions, and Jira or Linear integration. Gito is stateless: source code goes straight from the runner or local machine to the configured model provider and is not stored by Gito.

Key features

  • 01Reviews GitHub and GitLab pull requests through CI workflows and posts results as comments.
  • 02Supports OpenAI-compatible APIs, Anthropic, Google, and local LLM runtimes like Ollama and vLLM.
  • 03Runs `gito review` locally against a branch or reviews a remote repository.
  • 04Customizes prompts, output templates, and excluded files via a per-project TOML config.
  • 05Integrates with Jira and Linear for issue tracking.
  • 06Ships a Windows standalone installer and a `gito deploy` setup wizard.

Use cases

  • Adding automated first-pass review to every pull request in a GitHub repository.
  • Reviewing local changes before pushing, without sending code to a third-party review service.
  • Running code review with a self-hosted model in an air-gapped or privacy-sensitive environment.
  • Enforcing project-specific review standards through custom prompts in CI.

Quick start

  1. 1

    Install Gito from PyPI with pip

    pip install gito.bot
  2. 2

    Run the one-time setup wizard to configure your LLM connection

    gito setup
  3. 3

    Navigate to your repository root, switch to the branch to review, and run the review

    gito review

Things to know

  • Only Python 3.11 to 3.13 is listed as a prerequisite for the pip install route.
  • GitLab support is still in beta, and Bitbucket support is only planned.
  • Reviews require an LLM API key or a locally hosted model.

In collections

AI code review toolsAI coding assistants & agentsTesting frameworks & test automation tools

Topics

#code-review#llm#github-actions#static-analysis#ci-cd#python#developer-tools#ai#ai-code-analysis#ai-code-review

FAQ

What is Gito?
Open-source AI code reviewer that runs LLM-based reviews on GitHub pull requests and local branch changes, with any model provider. Gito is a command-line AI code reviewer that analyzes pull requests or local branch changes with a language model and reports likely security issues, bugs, and maintainability problems. It is built on ai-microcore, so it works with OpenAI-compatible APIs, Anthropic, Google, and local runtimes such as Ollama, vLLM, and llama.cpp.
Is Gito free to use? What license does it have?
Gito is open source under the MIT license. Check the license terms before using it in commercial products.
Is Gito still maintained?
Updated within 6 months. The last commit to Gito was on Oct 1, 2026.
What can I use Gito for?
Adding automated first-pass review to every pull request in a GitHub repository. Reviewing local changes before pushing, without sending code to a third-party review service. Running code review with a self-hosted model in an air-gapped or privacy-sensitive environment. Enforcing project-specific review standards through custom prompts in CI.
How do I install Gito?
Install Gito from PyPI with pip: pip install gito.bot Run the one-time setup wizard to configure your LLM connection: gito setup Navigate to your repository root, switch to the branch to review, and run the review: gito review
Where can I get Gito?
The source code and installation instructions are on GitHub: https://github.com/Nayjest/Gito

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

Related projects