Step 1 — Add your Anthropic API key​

Create an API key at platform.claude.com, then add it to your repo as a secret:

Settings → Secrets and variables → Actions → New repository secret, named ANTHROPIC_API_KEY.

GITHUB_TOKEN doesn't need to be added — GitHub provides it to workflows automatically.

Step 2 — Add the workflow file​

Create .github/workflows/code_review.yml in your repo:

name: Automated Code Review by Cody

on:
  pull_request:
    types: [opened, synchronize, reopened]

jobs:
  code_review:
    if: ${{ github.event.pull_request.head.repo.fork == false }}
    runs-on: ubuntu-latest
    permissions:
      contents: read
      pull-requests: write
    steps:
      - name: Checkout code
        uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8

      - name: Clone cody-code-reviewer repository
        run: |
          git clone https://github.com/codylabs/cody-code-reviewer.git
          cd cody-code-reviewer
          # Pin to a specific release commit to ensure stability https://github.com/codylabs/cody-code-reviewer/releases/tag/v1.3.0
          git checkout 42800f56034b775ea5dedfcdf44882213e17a70c

      - name: Set up Python 3.9
        uses: actions/setup-python@v4
        with:
          python-version: "3.9"

      - name: Install dependencies
        run: |
          python -m pip install --upgrade pip
          pip install -r requirements.txt

      - name: Run Code Review Model
        run: |
          python src/review_pull_request.py ${{ github.event.pull_request.number }} ${{ github.repository }} > ${{ github.workspace }}/output.txt
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
          MODEL: "claude-opus-4-8"

      - name: Comment on Pull Request
        run: |
          python src/comment_on_pr.py ${{ github.event.pull_request.number }} "${{ secrets.GITHUB_TOKEN }}" ${{ github.repository }}
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

The important line is MODEL: "claude-opus-4-8" — any claude-* model name routes the review to Anthropic. Prefer a cheaper review? Use claude-sonnet-5 or claude-haiku-4-5 instead.

Step 3 — Open a pull request​

That's it. On your next pull request, Cody fetches the diff, asks Claude to review it like a senior engineer, and posts the review as a PR comment — a summary of the change followed by focused feedback on functionality and security, with code suggestions in markdown.

Troubleshooting​

• No comment appears? Check the Actions run logs, and make sure the workflow has pull-requests: write permission (included in the yaml above).
• Authentication error? Confirm the ANTHROPIC_API_KEY secret name matches exactly and the key is active.
• Want reviews on GitLab or Azure DevOps? That's available in Cody Pro.

More details in the Quick Start Documentation.

Keywords: Claude code review GitHub Actions, Anthropic API GitHub Actions, automated PR review Claude, Claude Opus 4.8 tutorial, AI code review setup, GitHub Actions AI review workflow, CodyLabs, 2026

The short answer​

• Want the strongest reviews? Try claude-opus-4-8 — Anthropic's most capable Opus-tier model, particularly strong at spotting real bugs and explaining them clearly.
• Want great reviews at a lower price? claude-sonnet-5 delivers near-Opus quality on coding tasks.
• High-volume repos on a budget? claude-haiku-4-5 is the fastest, cheapest Claude option and still handles routine reviews well.
• Already invested in OpenAI? gpt-5 and the coding-focused gpt-5-codex remain fully supported — nothing changes for existing Cody users.

What matters in a code review model​

Finding real bugs, not nitpicks. The best review models flag functional and security issues and skip style bikeshedding. Cody's prompt explicitly asks for that focus regardless of which model you choose, but more capable models are noticeably better at separating signal from noise.

Reading large diffs in context. Claude's current models accept up to 1 million tokens of context, so even very large pull requests fit comfortably in a single review.

Cost per review. As of July 2026, Anthropic's list pricing per million tokens is roughly: Claude Opus 4.8 at $5 input / $25 output, Claude Sonnet 5 at $3 / $15 (with introductory pricing of $2 / $10 until the end of August 2026), and Claude Haiku 4.5 at $1 / $5. A typical pull request review uses a small fraction of a million tokens, so per-review costs are usually cents. OpenAI pricing varies by model — check their pricing page for current rates.

Try both in one afternoon​

Switching models in Cody is a one-line change:

env:
  ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
  MODEL: "claude-opus-4-8"   # or "claude-sonnet-5", "gpt-5", "gpt-5-codex"...

Run a few pull requests with each and compare the reviews on your own codebase — that beats any benchmark table.

• Setup Guide: Quick Start Documentation
• GitHub: https://github.com/codylabs/cody-code-reviewer

Keywords: Claude vs GPT-5, best AI model for code review, Claude Opus 4.8 vs GPT-5, AI code review comparison, automated code review GitHub, Claude Sonnet 5 code review, GPT-5 Codex review, CodyLabs, 2026

Same Cody, now on GitLab and Azure DevOps​

Cody Pro runs the same review engine as the free GitHub action — it reads your merge/pull request diff, reviews it like a senior engineer, and posts a summary plus focused feedback right on the request. It supports the same models, so you can review with Anthropic Claude (Opus 4.8, Sonnet 5, Haiku 4.5) or OpenAI GPT-5, selected with a single environment variable.

• GitLab — a drop-in CI job that runs on every merge request pipeline and posts the review as a note.
• Azure DevOps — a pipeline you add as a build validation policy; Cody posts the review as a comment thread on the pull request.

Like the free action, Cody Pro runs entirely in your own CI. Your code never passes through our servers — it goes straight from your pipeline to the AI provider you choose. Bring your own API key; there's nothing to host.

What's in the box​

Cody Pro is a one-time purchase and includes:

• Entry-point scripts for GitLab CI and Azure Pipelines
• Ready-made pipeline templates for both platforms
• The shared Claude/OpenAI model routing
• A setup guide and .env.example

Get it​

Get Cody Pro on Gumroad →

Still on GitHub? The free Cody action has you covered, and now supports Claude as well as OpenAI.

Pick your model​

Set MODEL in the template — any claude-* name uses your Anthropic key, anything else uses your OpenAI key.

Keywords: GitLab AI code review, Azure DevOps AI code review, automated merge request review, automated pull request review, GitLab merge request AI, Azure DevOps pull request bot, Claude code review, GPT-5 code review, CI/CD code review, Cody Pro, 2026

How it works​

Cody now routes your review to whichever provider matches the model you pick. Set the MODEL environment variable in your workflow: any model name starting with claude- is sent to Anthropic, and anything else goes to OpenAI. You only need the API key for the provider you choose.

Use Claude in your workflow​

1. Add ANTHROPIC_API_KEY as a repo secret (create a key at platform.claude.com).
2. Set MODEL in the "Run Code Review Model" step:

- name: Run Code Review Model
  run: |
    python src/review_pull_request.py ${{ github.event.pull_request.number }} ${{ github.repository }} > ${{ github.workspace }}/output.txt
  env:
    GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
    ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
    # Claude's most capable model for code review
    MODEL: "claude-opus-4-8"

That's it — Cody posts the Claude-written review straight onto your pull request.

Which Claude model should you pick?​

Existing workflows keep working unchanged — OPENAI_MODEL is still honoured and OpenAI models like gpt-5 and gpt-5-codex are fully supported.

• GitHub: https://github.com/codylabs/cody-code-reviewer
• Setup Guide: Quick Start Documentation

Keywords: Claude code review, Anthropic code review, Claude Opus 4.8 code review, Claude Sonnet 5, AI code reviewer GitHub Actions, automated code review, free AI code review, Claude vs GPT-5, CodyLabs, 2026

CodyLabs + GPT-5 Codex: Smarter Code Reviews​

OpenAI’s GPT-5 Codex brings a new level of intelligence and accuracy to code understanding. CodyLabs now lets you take full advantage of this powerful model for your automated code reviews.

How to Use GPT-5 Codex with CodyLabs​

It’s simple—just set the OPENAI_MODEL environment variable to gpt-5-codex in your workflow. For example:

- name: Run Code Review Model
  run: |
    python src/review_pull_request.py ${{ github.event.pull_request.number }} ${{ github.repository }} > ${{ github.workspace }}/output.txt
  env:
    GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
    OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
    # Use the new GPT-5 Codex model
    OPENAI_MODEL: "gpt-5-codex"

You can also choose any other model from the OpenAI model list if you prefer.

• GitHub: https://github.com/codylabs/cody-code-reviewer
• Setup Guide: Quick Start Documentation

Why CodyLabs?​

• Free and Open Source: No paywalls, no vendor lock-in—just great code reviews.
• Seamless Integration: Works with GitHub Actions and GitLab pipelines.
• Supports GPT-5 Codex: Get the most advanced code review AI available today.
• Your Workflow, Your Choice: Use GPT-5 Codex, GPT-4, or any other OpenAI model you like.

CodyLabs is the go-to solution for developers searching for a free, open source, AI-powered code review tool. Whether you want the best free AI code reviewer, an open source alternative to commercial tools, or to try GPT-5 Codex for code review, CodyLabs has you covered.

Keywords: free open source AI code reviewer, best free AI code review, open source GPT-5 Codex code review, automated code review, GitHub Actions, GitLab pipelines, CodyLabs, GPT5 Codex, GPT-5 Codex, GPT-4, GPT-5, AI code review tool, open source AI tools, 2025

What’s New with GPT-5?​

OpenAI has just released GPT-5, the most advanced language model yet, offering even deeper code understanding, smarter suggestions, and more accurate reviews. CodyLabs instantly integrates these improvements, giving you the best possible automated code review experience.

Why CodyLabs?​

• Free and Open Source: CodyLabs is 100% open source and free to use for everyone.
• AI-Powered Code Reviews: Get high-quality, automated code reviews powered by the latest AI models.
• Easy Integration: Works seamlessly with GitHub Actions and GitLab pipelines.
• No Vendor Lock-In: Your code, your workflow, your choice of AI model.

Get Started​

Ready to try the best free open source AI code reviewer?

• GitHub: https://github.com/codylabs/cody-code-reviewer
• Setup Guide: Quick Start Documentation

CodyLabs is the go-to solution for developers searching for a free, open source, AI-powered code review tool. Whether you’re looking for the best free AI code reviewer, an open source alternative to commercial tools, or just want to try GPT-5 for code review, CodyLabs has you covered.

Keywords: free open source AI code reviewer, best free AI code review, open source GPT-5 code review, automated code review, GitHub Actions, GitLab pipelines, CodyLabs, GPT5, AI code review tool, open source AI tools, 2025

Setting up Cody takes just a few minutes. Simply follow the introductory guide to install and integrate Cody into your environment.

Get started now by heading over to our setup docs!

Setting up Cody as part of your GitHub Actions or GitLab CI/CD pipelines is straightforward, thanks to the use of YAML configuration files. These files guide you through the setup process, making it easy to integrate Cody into your existing workflows. With this setup, Cody automatically reviews your code whenever you submit a pull request or merge request, providing insights and feedback at a fraction of the cost of traditional code review processes.

When you submit a Pull Request on GitHub, Cody automatically summarizes your changes and reviews your code with the expertise typical of a senior software engineer from a top tech company.

With our customizable template, you can select the model and configure the detail level of the responses.

Start Now!

Note that Cody Labs are not affiliated with any of the LLM providers.