The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of Python source code and configuration files, which is a known attack surface for indirect prompt injection.
Ingestion points: The skill's scope includes all Python files (**/*.py) and build configuration files (pyproject.toml, setup.py, requirements*.txt), which are ingested into the agent's context during code reviews or refactoring tasks.
Boundary markers: The instructions do not specify explicit delimiters or warnings to help the agent distinguish its own operational guidelines from instructions that might be embedded in code comments, docstrings, or metadata of the files being reviewed.
Capability inventory: The skill is intended to be used by agents with capabilities to read from and write to the file system to perform code reviews and refactoring.
Sanitization: No sanitization, filtering, or escaping is performed on the content of the ingested files before they are processed by the agent.
[SAFE]: The skill recommends the use of vendor-owned and reputable third-party development tools. It suggests pyscn, a static analysis tool hosted on the author's GitHub repository, and other well-known tools like ruff, mypy, and pytest for automated quality enforcement.

coding-standards