The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local Python scripts to perform its tasks. The test suite (tests/test_doc_helper_scripts.py) utilizes subprocess.run() to execute these internal scripts (e.g., index_docs.py, build_topic_map.py) for validation purposes. These calls are restricted to the skill's own bundled scripts and are used for integration testing.
[DATA_EXPOSURE]: The skill's instructions explicitly include a security-positive rule (Section 5.7) directing the agent to redact or remove secrets like API keys or tokens if they are encountered in documentation files. It uses placeholders for examples, following secret management best practices.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the repository's Markdown files. However, its operations are limited to structural formatting and metadata indexing. It does not evaluate or execute the content of the documentation as instructions, and the capabilities available to the scripts (file read/write) are scoped to the intended documentation management purpose.

documentation