zyx77550/Sparda Security Gate
The trust layer for AI code: prove a deploy, diff PR behavior, and gate AI edits. 100% local, deterministic, no keys.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| mode | 'apocalypse' (prove the deploy, default), 'review' (comment the PR behavior diff), or 'prove' (comment the trust verdict + badge) | no | apocalypse |
| working-directory | Directory of the app (where package.json lives) | no | . |
| version | sparda-mcp version to run (npm dist-tag or semver) | no | latest |
| sarif | apocalypse: also emit .sparda/apocalypse.sarif and upload it to code scanning | no | false |
| base-ref | review: base git ref to diff against (default: the PR base branch) | no | "" |
| comment | review: post/update a sticky PR comment with the behavior diff | no | true |
| fail-on-severity | review: fail the check on findings — 'none' (default, comment-only), 'high', or 'critical' | no | none |
Outputs
no outputs