bai-os/Sentinel Review
AI code review with local agents (Claude Agent SDK): grounded findings (file + line) posted as a single PR review. Code goes only to the model provider you configure — no third-party review service.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| anthropic-api-key | Anthropic API key for the review agents. Store it as a repository or organization secret. Absent on fork PRs by design — the review is skipped with an explanatory job summary. | no | "" |
| github-token | Token used to read the PR and post the review. | no | ${{ github.token }} |
| agents | Comma-separated subset of reviewers to run (correctness|security|tests), overriding the repository policy. Leave empty to honor .sentinel.yml. | no | "" |
| dry-run | Print the review into the job summary instead of posting to the PR. | no | false |
| base-url | GitHub API base URL override (GitHub Enterprise Server). | no | "" |
Outputs
no outputs