dast-scanning
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a documentation resource for using legitimate security tools. It contains no executable malicious code and follows best practices for authorization and confirmation gates.- [EXTERNAL_DOWNLOADS]: References official Docker images from the GitHub Container Registry (ghcr.io/zaproxy/zaproxy) and official GitHub Actions (zaproxy/action-full-scan). These are well-known, trusted sources for these security tools.- [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection via the processing of external application data during security scans.
- Ingestion points: External URLs and hosts scanned by tools, and tool-generated reports (e.g., full-report.html).
- Boundary markers: Includes a 'Mandatory confirmation gate' requiring target verification and explicit authorization before any active steps.
- Capability inventory: Command execution via Docker, network requests via Python requests, and file writing for reporting.
- Sanitization: Instructions advise manual review of findings and coordination with development teams, though programmatic sanitization of tool output is not specified.
Audit Metadata