manage-skills

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • COMMAND_EXECUTION (SAFE): The skill invokes local Python scripts (audit_skills.py and check_compliance.py) to perform its core auditing functions. These scripts are part of the skill's own package and are used for local file system operations.
  • Indirect Prompt Injection (LOW): The skill monitors a directory for third-party skill files, creating an ingestion surface. Evidence: 1. Ingestion point: Files within the .claude/skills/ directory. 2. Boundary markers: No explicit delimiters or ignore-instructions mentioned in the markdown. 3. Capability: Executes local Python scripts via shell commands. 4. Sanitization: Verification of script safety cannot be determined from the markdown alone, but the primary task is structural validation rather than text-based instruction processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 01:07 AM