mr8bit/Skills Verified — AI Agent Trust Scanner
Scan an AI agent repository for vulnerabilities and compute a Trust Score.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| source | Path or URL to scan | no | . |
| threshold | Minimum score (0-100); fail if below | no | "" |
| threshold-grade | Minimum grade (A/B/C/D/F); fail if worse | no | "" |
| skip | Comma-separated list of analyzers to skip | no | "" |
| only | Comma-separated list of analyzers to run exclusively | no | "" |
| use-docker | Use Docker image instead of pip install | no | false |
| python-version | Python version to use (if not Docker) | no | 3.11 |
| comment-on-pr | Post Markdown comment on PR | no | true |
| comment-style | Markdown comment style: full or summary | no | full |
| generate-badge | Generate badge.json artifact | no | true |
Outputs
| name | description |
|---|---|
| score | Overall trust score (number 0-100) |
| grade | Overall trust grade (letter A-F) |
| passed | Whether the scan passed the threshold (true/false) |
| report-path | Path to the JSON report file |