skill-creator

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • [PROMPT_INJECTION] (SAFE): No instruction bypass, safety filter overrides, or adversarial role-play patterns were detected in the skill instructions.
  • [DATA_EXFILTRATION] (SAFE): Python scripts utilize standard libraries for local operations and do not include any network-capable code or access to sensitive system paths.
  • [COMMAND_EXECUTION] (LOW): The skill instructions facilitate the execution of included helper scripts for initialization and packaging. These scripts use the pathlib and zipfile modules for standard file management.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill ingests user input to generate skill templates. 1. Ingestion points: User input for skill purpose and metadata. 2. Boundary markers: Absent. 3. Capability inventory: init_skill.py (file creation) and package_skill.py (directory zipping). 4. Sanitization: quick_validate.py implements regex-based naming checks and filters angle brackets in descriptions.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 01:16 PM