w3osc/skill-warden
Scan AI skills for security threats - prompt injection, jailbreaks, secret grabbing, and more
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| target | GitHub URL or local path to skill(s) to scan. Defaults to the repository root. | no | ${{ github.workspace }} |
| output-format | Output format: pretty, json, sarif | — | sarif |
| sarif-file | Path to write SARIF output file | — | skill-warden-results.sarif |
| fail-on-advisory | Fail the action if advisory violations are found | — | false |
| github-token | GitHub token for private repositories | — | ${{ github.token }} |
| upload-sarif | Upload SARIF results to GitHub Security tab | — | true |
| no-quality | Skip quality checks | — | false |
| no-ai-score | Skip AI slop scoring | — | false |
Outputs
| name | description |
|---|---|
| hard-passed | Whether all hard security checks passed (true/false) |
| has-advisories | Whether advisory violations were found (true/false) |
| sarif-file | Path to the SARIF output file |