python-tooling
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process common Python project files which represents an indirect prompt injection surface if these files are authored by a malicious actor.\n
- Ingestion points: Reads pyproject.toml, requirements.txt, pytest.ini, ruff.toml, mypy.ini, uv.lock, and poetry.lock (defined in SKILL.md metadata).\n
- Boundary markers: None explicitly defined in the instructions to distinguish between trusted rules and untrusted file content.\n
- Capability inventory: Executes subprocesses for ruff, pyright, pytest, and python -m compileall.\n
- Sanitization: Relies on the security of the underlying tools and does not perform independent sanitization of file content before processing.\n- [DYNAMIC_EXECUTION]: The skill workflow involves executing Python code via pytest and compiling source trees using the compileall module.\n
- Evidence: Instructions in SKILL.md and references/tooling-gates.md to run 'pytest' and 'python -m compileall'.\n- [COMMAND_EXECUTION]: The skill facilitates the execution of CLI tools to perform linting and security audits on the local environment.\n
- Evidence: Verification workflow in SKILL.md specifies running 'ruff check', 'pyright', and 'pip-audit'.
Audit Metadata