anufdo/AI Code Review Action
AI-powered code review for pull requests with inline comments, security, performance, and best-practices analysis
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| github-token | GitHub token for API access | yes | ${{ github.token }} |
| openai-api-key | OpenAI API key for AI-powered code review | no | — |
| anthropic-api-key | Anthropic API key for AI-powered code review (alternative to OpenAI) | no | — |
| openrouter-api-key | OpenRouter API key for AI-powered code review (supports multiple models) | no | — |
| ai-provider | AI provider to use (openai, anthropic, openrouter) | no | openai |
| model | AI model to use for code review. For OpenRouter: use model names like 'anthropic/claude-3-sonnet', 'openai/gpt-4', 'meta-llama/llama-3-70b-instruct', etc. | no | gpt-4 |
| openrouter-base-url | OpenRouter base URL (usually https://openrouter.ai/api/v1) | no | https://openrouter.ai/api/v1 |
| max-files | Maximum number of files to review | no | 20 |
| exclude-patterns | Comma-separated list of file patterns to exclude from review | no | *.lock,*.min.js,*.map,node_modules/**,dist/**,build/** |
| review-level | Review level (basic, standard, detailed) | no | standard |
| language-hints | Programming languages in the project (auto-detected if not provided) | no | — |
| custom-prompts | Custom review prompts (JSON format) | no | — |
| enable-security-review | Enable security-focused code review | no | true |
| enable-performance-review | Enable performance-focused code review | no | true |
| enable-best-practices | Enable best practices review | no | true |
Outputs
| name | description |
|---|---|
| review-summary | Summary of the code review |
| issues-found | Number of issues found |
| files-reviewed | Number of files reviewed |
| review-url | URL to the review comment |