raye-deng/Open Code Review
AI code quality gate — detect hallucinations, stale APIs, logic gaps in PRs
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| sla | Scan level: L1 (structural), L2 (+ embedding), L3 (+ LLM deep scan) | no | L1 |
| threshold | Minimum score to pass (0-100) | no | 70 |
| scan-mode | Scan mode: diff (changed files only) or full | no | diff |
| ollama-url | Ollama API URL for local AI (L2/L3). If not provided, L2 uses TF-IDF fallback. | no | — |
| fail-on-low-score | Fail workflow if score is below threshold | no | true |
| github-token | GitHub token for PR comments | no | ${{ github.token }} |
Outputs
| name | description |
|---|---|
| score | Quality score (0-100) |
| grade | Letter grade (A-F) |
| issues-count | Total issues found |