web-security-audit
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary purpose is defensive security auditing and education; it contains no malicious instructions or behavior overrides.
- [SAFE]: The skill references legitimate, well-known security tools (e.g., bandit, semgrep, sqlmap) and follows industry best practices for vulnerability identification and remediation.
- [PROMPT_INJECTION]: The skill identifies user-supplied code snippets as the primary input for analysis, creating an attack surface for indirect prompt injection. Maliciously crafted code could contain instructions intended to distract the agent or bias its security verdict.
- Ingestion points: User-provided code snippets or file fragments as specified in the SKILL.md instructions.
- Boundary markers: The instructions do not define specific delimiters or "ignore previous instructions" warnings to isolate user code from the audit logic.
- Capability inventory: The skill is strictly instructional and diagnostic; it does not request tool access, network capabilities, or file system write permissions that could be abused via injection.
- Sanitization: No input validation or sanitization of the provided code is described in the workflow.
Audit Metadata