skill-creator

Warn

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PRIVILEGE_ESCALATION]: The script scripts/init_skill.py applies executable permissions (chmod 0o755) to dynamically created files like example.py.
  • [DYNAMIC_EXECUTION]: The skill programmatically generates executable Python scripts and documentation from internal templates using user-supplied names and paths.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided skill names and paths that are interpolated into file system operations and templates without sufficient sanitization.
  • Ingestion points: Command-line arguments in scripts/init_skill.py and scripts/package_skill.py.
  • Boundary markers: Absent in the interpolation of templates and path construction.
  • Capability inventory: Directory creation, file writing, permission modification, and ZIP creation.
  • Sanitization: Input strings are used in path resolution without validation, allowing potential absolute path injection or traversal via the skill name argument.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 22, 2026, 07:32 AM