🤖 AI Code Review

What is AI Code Review?

It is an Artificial Intelligence-powered tool designed to automate and enhance the code review process. It supports GitLab and GitHub, and offers flexibility with AI providers like Gemini, Anthropic, and local models via Ollama.

Our goal is to assist in the review task by providing deep analysis and rich context, making human reviews more effective and focused on critical logic. It also provides a valuable safety net for individual developers who lack regular peer review.

Read the Full User Documentation

Three Primary Workflows

🤖

CI Integration

Automated reviews in your CI/CD pipeline.

🌐

Remote Reviews

Analyze existing MRs/PRs from the terminal.

🔍

Local Reviews

Review your changes before committing.

Primary Use Cases

Explore the three main workflows of the tool. Click on each tab to see the details and code examples.

Integration into the CI/CD Pipeline

This is the core use case. The tool integrates directly with GitLab and GitHub to provide automated reviews on every Merge Request or Pull Request. The key command is --post, which generates the review and publishes it as a comment.

Example: GitLab CI/CD (.gitlab-ci.yml)

# Runs only on Merge Request events
ai-code-review:
  stage: code-review
  image: registry.gitlab.com/redhat/edge/ci-cd/ai-code-review:latest
  script:
    - ai-code-review --post
  rules:
    - if: '$CI_PIPELINE_SOURCE == "merge_request_event"'

Example: GitHub Actions (ai-review.yml)

# Runs in a container for easy setup
# Requires: GITHUB_TOKEN and AI_API_KEY secrets
jobs:
  ai-review:
    permissions:
      pull-requests: write # IMPORTANT
    container:
      image: registry.gitlab.com/redhat/edge/ci-cd/ai-code-review:latest
    steps:
      - name: Run AI Review
        env:
          GITHUB_TOKEN: ${secrets.GITHUB_TOKEN}
          AI_API_KEY: ${secrets.GEMINI_API_KEY}
        run: ai-code-review --post

The result is a detailed comment on the MR/PR with an executive summary and the technical code analysis.

Remote Reviews (Terminal Analysis)

Analyze existing Merge Requests or Pull Requests directly from your terminal without posting comments. This is useful for manual verification or saving the review to a local file.

Examples: Analyze and Post

GitLab MR Analysis

# Analyze (Prints to terminal)
ai-code-review group/project 123

# Post review as a comment
ai-code-review group/project 123 --post

GitHub PR Analysis

# Analyze (Prints to terminal)
ai-code-review --platform github owner/repo 456

# Post review as a comment
ai-code-review --platform github owner/repo 456 --post

Local Changes Review (Pre-commit)

Get an instant review of your uncommitted local changes. This is an excellent way to ensure code quality before sharing it with your team.

Example: Local Review

# Review current changes vs main branch
ai-code-review --local

# Or against a different target branch
ai-code-review --local --target-branch develop

Advanced Features

Discover the functionalities that make this tool more powerful and flexible. Click on each card for more details.

Local LLM Usage

Utilize local models (Ollama) for full privacy and cost savings.

Context Files

Improve review quality by providing project architecture context to the AI.

Context Generation

Create context files automatically by analyzing your Git repository.

Context7 Enrichment

Add official library documentation to the context for accurate reviews.

Smart Skip Review

Automatically skip unnecessary reviews (dependabot, WIPs) to save costs.

CI/CD Documentation

Fetch recent changes and deprecations from official CI/CD documentation for accurate reviews.