owasp-zap

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and executes tools from the official owasp/zap2docker-stable Docker image for security scanning tasks.
  • [COMMAND_EXECUTION]: Runs local shell commands to execute security scans via Docker (zap-baseline.py, zap-full-scan.py, zap-api-scan.py) and the zaproxy binary.
  • [INDIRECT_PROMPT_INJECTION]: As a web security tool, it ingests data from external URLs and APIs. This surface could be targeted with malicious content, though the skill provides instructions for scoping and manual validation to mitigate risks.
  • Ingestion points: Reads content from user-provided target URLs and OpenAPI specifications (SKILL.md).
  • Boundary markers: Explicitly instructs users to scope scanning to approved application areas and manually validate findings.
  • Capability inventory: Uses docker run and shell commands to perform network-based security analysis (SKILL.md).
  • Sanitization: Not applicable for the static definition, but the procedure recommends manual validation of findings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 02:38 AM
Security Audit — agent-trust-hub — owasp-zap