koach08/AgentSift Security Scan
Scan AI agent plugins, MCP packages, and ClawHub skills for security issues
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| targets | Space-separated list of targets to scan (e.g., 'npm:@mcp/server-x clawhub:skill-y') | no | "" |
| target-file | Path to a file containing targets (one per line) | no | "" |
| path | Local directory path to scan | no | "" |
| fail-on | Fail if findings at or above this severity (critical, high, medium, low) | no | high |
| rules | Path to custom YAML rules directory | no | "" |
| format | Output format (human, json, sarif) | no | sarif |
| sarif-file | Path to write SARIF output | no | agentsift-results.sarif |
Outputs
| name | description |
|---|---|
| risk-score | Highest risk score across all scanned targets |
| findings-count | Total number of findings |
| sarif-file | Path to SARIF output file |