skill-creator

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The scripts/init_skill.py script performs dynamic file system operations, including creating directories and writing a Python script to disk based on user-provided names.\n- [COMMAND_EXECUTION]: The scripts/init_skill.py script uses chmod(0o755) to set executable permissions on a dynamically generated script file (scripts/example.py).\n- [PROMPT_INJECTION]: The skill provides an interface for generating code based on untrusted user input, creating a surface for indirect prompt injection.\n
  • Ingestion points: The <skill-name> and <path> parameters passed to scripts/init_skill.py.\n
  • Boundary markers: None; user-supplied strings are directly formatted into Python code and Markdown templates.\n
  • Capability inventory: File system writing and permission modifications within scripts/init_skill.py and archiving in scripts/package_skill.py.\n
  • Sanitization: No validation or path-cleaning is performed on the skill_name input before using it to create paths or populate code content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 11:43 PM