theoverengineer/over-review
AI-powered Pull Request reviewer using your own model
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| llm-model | LLM model to use for review (e.g., gpt-4o, claude-3.5-sonnet) | yes | — |
| llm-api-key | API key for the LLM provider | yes | — |
| llm-base-url | Base URL for the LLM API (optional, for OpenAI-compatible APIs) | no | — |
| llm-provider | LLM provider (ai-sdk only in V1) | no | ai-sdk |
| review-mode | Review mode: auto, manual, or cli | no | auto |
| full-mode | Force full review instead of incremental | no | false |
| llm-structured-outputs | Enable structured JSON outputs (set to false for custom OpenAI-compatible endpoints without schema support) | no | true |
| style-guide-rules | Additional repository-specific review guidance text appended to the review prompt | no | — |
| llm-timeout-ms | Timeout in milliseconds for LLM requests (0 to disable) | no | 30000 |
| github-api-url | GitHub API URL (for GitHub Enterprise or custom endpoints) | no | — |
| github-server-url | GitHub server URL (for GitHub Enterprise) | no | — |
| github-token | GitHub token for API authentication | no | — |
| debug | Enable debug logging | no | false |
| dry-run | Skip GitHub write operations | no | false |
Outputs
no outputs