AI Code Reviewer
AI Code Reviewer automatically reviews GitHub pull requests and posts a summary with inline findings on changed lines.
What AI Code Reviewer is
AI Code Reviewer automatically reviews GitHub pull requests and posts a summary with inline findings on changed lines. Findings can include severity, explanations, and one-click suggestion blocks. It runs as a GitHub Action configured through a workflow file, without a server or checkout step. Model options include Anthropic, OpenAI, local Ollama, and OpenAI-compatible endpoints. Review settings cover severity thresholds, file inclusion and exclusion patterns, incremental reviews, and Markdown rules for house style or framework conventions. The action reads pull-request diffs through the GitHub API and does not check out, build, or run pull-request code. Provider keys are passed through GitHub secrets and masked in logs. The project says it does not proxy requests through a service it runs; with Ollama on a self-hosted runner, code and diffs stay within the user's network. It is MIT-licensed and free, though model-provider usage may cost money. By default, reviews are skipped when changes exceed 50 files, and comments are capped at 20 per pull request. Reviews of public-repository fork pull requests are skipped because those workflows cannot access repository secrets.
Good
- Supports Anthropic, OpenAI, and Ollama.
- Custom Markdown review rules are supported.
- Does not check out or run pull-request code.
- Ollama can keep diffs within the user's network.
Watch out
- Public-repository fork reviews are skipped.
- Reviews skip changes over 50 files by default.
- Inline comments are capped at 20 by default.
Who it suits
This free, MIT-licensed project suits teams looking for self-hosting, a choice of model providers, and control over code privacy. It is intended for teams using GitHub pull requests and Actions.
Verdict
AI Code Reviewer offers configurable automated reviews within a GitHub Action workflow, with model choice and self-hosting options. Its default file and comment limits, plus the restriction on public fork reviews, are important workflow constraints.
Plans and prices
From AI Code Reviewer's own pricing page, checked 21 Oct 2026.
| Plan | Price | What it gives you |
|---|---|---|
| Free | Free | Self-hosted GitHub Action · no per-seat charge · model usage billed separately · Free; users pay only their model-provider or compute costs |
The facts
Each one read from AI Code Reviewer's own pages, with the date it was checked.
What it does
- Chrome extension details
- The unrelated Chrome listing reports eight users, version 0.0.5, 346 KiB size and an update date of February 21, 2026.checked 21 Oct 2026
- Configuration
- Settings can be supplied in .github/ai-code-reviewer.yml or as action inputs, with action inputs taking precedence.checked 21 Oct 2026
- Custom guidelines
- A Markdown rules file can provide house style or framework conventions for reviews.checked 4 Oct 2026
- Custom rules
- A Markdown rules_file can add repository-specific coding guidelines, and it is read from the base branch.checked 21 Oct 2026
- Data handling
- The project says it does not proxy requests through a service it runs; with Ollama on a self-hosted runner, code and diffs stay within the user’s network.checked 4 Oct 2026
- Development quality bar
- Contributor documentation specifies type checking, linting, Vitest tests, action builds and at least 80% coverage for all metrics.checked 21 Oct 2026
- Execution safety
- The action reads diffs through the GitHub API and does not check out, build or execute pull-request code.checked 21 Oct 2026
- Incremental behavior
- When incremental mode is enabled, a new push reviews only newly added commits and rerunning the same head SHA is skipped.checked 21 Oct 2026
- Installation
- Installation consists of adding a GitHub Actions workflow using xiaofuqing13/ai-code-reviewer@v1 and a provider secret; checkout and a separate server are not required.checked 21 Oct 2026
- Integrations
- Documented OpenAI-compatible examples include DeepSeek, GLM, Kimi, Together, Groq, vLLM and LM Studio through a configurable base URL.checked 21 Oct 2026
- Large-change handling
- For very large pull requests the reviewer caps files or chunks and reports the truncation in its summary.checked 21 Oct 2026
- Licence
- The repository is released under the MIT License, permitting reuse, modification, distribution and sale subject to retaining the notice and disclaiming warranty.checked 21 Oct 2026
- License and support
- The project is MIT-licensed and welcomes contributions; the README does not state a paid support offering.checked 4 Oct 2026
- Limit: G2 CodeAnt listing
- Reviews on that separate CodeAnt product mention stylistic nitpicks, cautious suggestions requiring manual adjustment and integration taking time.checked 21 Oct 2026
- Limit: Product Hunt name collision
- Product Hunt discussion for the unrelated product reports single-file or text-box-only review, missed async/await errors, client-side exceptions and long-code formatting problems.checked 21 Oct 2026
- Model providers
- The action supports Anthropic Claude, OpenAI, local Ollama and OpenAI-compatible endpoints.checked 21 Oct 2026
- Notable limits
- The default configuration skips reviews when more than 50 files change and caps inline comments at 20 per pull request.checked 4 Oct 2026
- Package status
- The package manifest describes version 0.1.0, marks the workspace private and identifies the project as an MIT-licensed self-hosted GitHub pull-request reviewer.checked 21 Oct 2026
- Permissions
- The example workflow requests contents read and pull-requests write permissions.checked 21 Oct 2026
- Praise: G2 CodeAnt listing
- Reviews on the separately identified CodeAnt product praise accurate actionable findings, faster reviews and catching issues missed by linters.checked 21 Oct 2026
- Privacy claim
- The project says it operates without a vendor proxy and that Ollama on a self-hosted runner keeps code and diffs inside the user network.checked 21 Oct 2026
- Private repositories
- The FAQ says private repositories can run reviews for collaborator branches because their workflows can access secrets.checked 21 Oct 2026
Money
- Cost example
- The FAQ estimates roughly $0.05–$0.20 per typical pull request with Claude Sonnet, a few cents with Haiku and zero model charge with local Ollama apart from compute.checked 21 Oct 2026
The company
- Maker identity
- The GitHub account displays the name xiaofuqing and lists 48 repositories, six followers and one following.checked 21 Oct 2026
Alternatives to AI Code Reviewer
- 1
DeepSource
Free planWebAPILinuxmacOS8.0From $30 a month - 2
Cubic
Free planWebAPILinuxmacOS7.9From $30 a month - 3
Sourcery
Free planWebAPIExtensionLinux7.9From $12 a month - 4
Ellipsis
Free planWebAPILinuxmacOS7.8Free plan - 5
Greptile
Free planWebAPIExtensionLinux7.7From $30 a month - 6
CodeAnt AI
Free planWebAPIExtensionSelf-hosted7.5From $20 a month