skill-creator

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute local Python utility scripts (scripts/init_skill.py and scripts/quick_validate.py) to automate the creation and validation of new skill folders. These scripts use standard libraries (pathlib, re, sys) and are restricted to managing files within the repository's .github/skills/ directory.
  • [PROMPT_INJECTION]: The skill handles an inherent risk of indirect prompt injection as its primary purpose is to generate instructions for other AI agents based on user input. However, the skill incorporates a multi-phase workflow (Phase 4 and 5) that includes automated validation and manual quality checklists to mitigate the risk of generating malformed or unsafe instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 12:06 AM
Security Audit — agent-trust-hub — skill-creator