vanessasvictoria/PR Review Bot
Deterministic checks + LLM-assisted PR review with sticky comments
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 (defaults to github.token) | no | ${{ github.token }} |
| openai_api_key | OpenAI API key for LLM review (optional, skipped if not provided) | no | — |
| model | LLM model identifier (e.g., gpt-4o-mini) | no | gpt-4o-mini |
| max_diff_bytes | Maximum total diff size in bytes | no | 200000 |
| max_files | Maximum number of files to process | no | 200 |
| include_file_context | Include full file content for small files | no | false |
| file_context_max_bytes | Maximum file size to include full content | no | 20000 |
| risk_rules_path | Path to risk rules YAML file (optional) | no | .pr-review-bot/risk_rules.yml |
| sticky_comment_tag | Tag for identifying sticky comments | no | PR_REVIEW_BOT |
| fail_on_high_risk | Fail workflow if risk score exceeds threshold | no | false |
| high_risk_threshold | Risk score threshold (0-10) | no | 8 |
| enable_dependency_audit | Run dependency audit checks | no | true |
| enable_tests | Run test suite | no | true |
| enable_lint | Run linting checks | no | true |
| log_level | Logging level (DEBUG, INFO, WARNING, ERROR) | no | INFO |
| redact_patterns | Comma-separated regex patterns for redaction | no | — |
| extra_instructions | Additional instructions for LLM review | no | — |
Outputs
no outputs