skill-description-audit
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze untrusted SKILL.md files (Ingestion point). While it acknowledges the injection surface in its instructions, it lacks explicit prompt boundary markers to isolate the analyzed content (Boundary markers: absent). Given its capabilities to generate reports and execute local scripts (Capability inventory), it presents a low-severity indirect injection risk where malicious instructions in audited files could attempt to influence the agent.
- [COMMAND_EXECUTION]: The skill requires the agent to execute local shell and Python commands, specifically 'python3 skill-audit-fp.py', 'bash skill-smoke.sh', and 'skill-executability-smoke.py', to perform content hashing and verification tasks during the audit workflow.
- [DYNAMIC_EXECUTION]: The skill relies on the execution of external Python and Bash scripts at runtime to calculate audit fingerprints and perform smoke testing, representing a localized dynamic execution pattern.
Audit Metadata