suchithnarayan/AI Skill Security Scanner
Scan AI agent skills and plugins for security vulnerabilities on pull requests
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| ai-provider | AI provider to use (openai, gemini, bedrock, anthropic, azure) | no | openai |
| ai-model | Specific AI model to use (defaults to provider default) | no | — |
| fail-on | Severity threshold to fail the check (critical, high, medium, low) | no | high |
| scan-mode | Scan mode (strict, balanced, permissive) | no | balanced |
| static | Also run static rule-based analysis alongside AI scan | no | false |
| workers | Max parallel LLM calls | no | 4 |
| triage-threshold | AI triage confidence threshold (0.0-1.0) | no | 0.5 |
Outputs
| name | description |
|---|---|
| verdict | Scan verdict (pass or fail) |
| new-findings | Number of new vulnerabilities found |
| risk-delta | Overall risk change (increased, decreased, unchanged) |