skill-scanner

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the uv run command to execute a local Python script (scripts/scan_skill.py) for automated security analysis. This is a standard and justified use of command execution for a security tool that performs file-based scanning.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an indirect prompt injection surface because its primary function is to ingest and analyze untrusted content from other skills.
  • Ingestion points: The agent reads the SKILL.md, scripts, and directory contents of third-party skills provided by the user.
  • Boundary markers: The instructions establish a multi-phase workflow (Phases 1-8) to separate discovery, automated scanning, and manual analysis, though it lacks explicit technical delimiters for external content.
  • Capability inventory: The skill has access to directory listing (ls) and script execution (uv run).
  • Sanitization: The skill incorporates defensive instructions, explicitly warning the agent to distinguish between security documentation (legitimate) and active exploits (malicious), and provides structured reference patterns to guide the evaluation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 07:21 AM