The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill's workflow includes steps to validate documentation examples using python -m doctest and pytest --doctest-modules. This process involves the execution of code snippets extracted directly from the docstrings of files being documented.
[COMMAND_EXECUTION]: The agent is instructed to run several shell commands for linting and type-checking, such as tsc --noEmit and npx @redocly/cli lint. It also suggests the use of Python tools like pydocstyle and interrogate for measuring documentation coverage.
[EXTERNAL_DOWNLOADS]: The reference guides recommend installing various third-party utility packages and dependencies using npm and pip, including eslint-plugin-jsdoc, linkchecker, and broken-link-checker for site validation and documentation linting.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code and API specifications to generate its documentation output, creating an attack surface for indirect prompt injection.
Ingestion points: The skill reads source files (.py, .ts, .js) and API definitions (YAML, JSON) provided by the user.
Boundary markers: There are no explicit markers used to separate the content being documented from the instructions given to the agent.
Capability inventory: The skill can read/write files, execute shell commands, and run code snippets via the Python doctest module.
Sanitization: The skill does not appear to perform any sanitization or validation of the input content before it is used for documentation generation or validation tasks.

code-documenter