security-audit-static
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, command injections, or exfiltration vectors were identified in the skill's instructions. The toolset is correctly restricted to read-only git operations and localized report writing.
- [INDIRECT_PROMPT_INJECTION]: The skill is specifically designed to analyze untrusted third-party repositories, which is an inherent risk surface for indirect prompt injection. However, the instructions provide robust mitigations by requiring the agent to treat all repository content as data for analysis only and to specifically flag any content attempting to 'steer' the auditor as a security finding. The process requires mandatory verbatim citation verification and a 'self-refute' phase to ensure the integrity of results against adversarial content in the audited codebase.
- [COMMAND_EXECUTION]: The skill uses shell tools but restricts them to safe git subcommands (log, diff, show, status) for repository inspection, ensuring no arbitrary code execution is possible.
Audit Metadata