mikecavallo/ClawGuard Security Scanner
Scan AI agent skills for security threats — static analysis, dependency scanning, prompt injection detection
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Feb 16, 2026
- License
- None
Pinned Snippet
uses: mikecavallo/clawguard@ca722f64834eff289eefac8765c147b9d0dae918 # v1.0.2tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| path | Path to the skill directory to scan | yes | . |
| severity-threshold | Minimum severity to fail the action (critical, high, medium, low) | no | high |
| output-format | Output format (md, json) | no | md |
| semantic | Enable LLM-powered semantic analysis (requires api-key) | no | false |
| api-key | API key for semantic analysis (use secrets!) | no | — |
| node-version | Node.js version to use | no | 20 |
Outputs
| name | description |
|---|---|
| risk-level | Risk level (SAFE, LOW, MEDIUM, HIGH, CRITICAL) |
| risk-score | Risk score (0-100) |
| findings-count | Total number of findings |
| report | Full scan report |