skill-authoring

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides standard instructions for executing local Python utility scripts within the project repository.
  • Evidence: Commands include python scripts/scaffold.py, python scripts/validate_skill.py, python scripts/audit_skill_system.py, and python scripts/bundle.py used for maintenance tasks.
  • [PROMPT_INJECTION]: The skill acts as an authoring and auditing tool, creating a surface for indirect prompt injection when processing untrusted third-party skill files.
  • Ingestion points: The validation and auditing tools (validate_skill.py, audit_skill_system.py) read external skill definitions from the file system.
  • Boundary markers: The skill documentation does not explicitly specify delimiters or sanitization logic for the content being audited.
  • Capability inventory: The skill has the capability to execute local scripts and bundle files into zip archives.
  • Sanitization: Not explicitly mentioned in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 09:10 PM
Security Audit — agent-trust-hub — skill-authoring