skill-creator-linter
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Python script (
lint.py) to process and validate skill metadata. This is standard behavior for an internal linter tool.\n- [PROMPT_INJECTION]: The skill processes potentially untrusted meta-skill definitions provided through standard input, representing an indirect injection surface.\n - Ingestion points: The
with.skill_mdparameter passes raw skill content to the linter's stdin.\n - Boundary markers: None identified for the input stream.\n
- Capability inventory: The tool writes to temporary directories and invokes internal parsing and scheduling logic from the
opensquillalibrary.\n - Sanitization: The linter actively enforces safety rules (e.g., G1.6) requiring sanitization filters in the skills it processes, enhancing the overall security of the skill ecosystem.\n- [SAFE]: The script uses dynamic path derivation to import internal dependencies from the
opensquillapackage based on its own installation directory. This is a common and acceptable practice for integrated development tools within a specific framework.
Audit Metadata