skill-create

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/create_skill.py creates directories and writes files to the local filesystem. It uses the pathlib module to manage directory creation and file writing for the new skill structure.\n- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it interpolates user-provided metadata into the instructions of the generated SKILL.md file.\n
  • Ingestion points: User input for skill name, description, author, and tags are ingested via CLI arguments in scripts/create_skill.py.\n
  • Boundary markers: No delimiters are used to wrap interpolated metadata in the generated markdown files.\n
  • Capability inventory: The skill is restricted to local filesystem operations and lacks network or sensitive data access capabilities.\n
  • Sanitization: The name parameter is validated with a strict regex (^[a-z0-9-]+$), though other metadata fields are not sanitized before interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 06:35 PM