The Agent Skills Directory

[SAFE]: The skill functions as a technical documentation assistant. All analyzed instructions and reference materials are legitimate and focused on software documentation best practices.
[EXTERNAL_DOWNLOADS]: The reference files mention several industry-standard tools for documentation generation and linting, such as Docusaurus, MkDocs, pydocstyle, and interrogate. These references are informative and point to trusted packages and repositories.
[PROMPT_INJECTION]: Indirect prompt injection surface analysis:
Ingestion points: The skill processes code files and API specifications provided by the user (as defined in the workflow in SKILL.md).
Boundary markers: No explicit boundary markers or 'ignore' instructions are defined for the input content.
Capability inventory: The skill primarily generates text and documentation; no dangerous system capabilities, subprocess calls, or arbitrary network operations are present in the scripts.
Sanitization: No explicit sanitization of input content is described. Given the purely descriptive nature of the skill and lack of hazardous capabilities, the overall risk is minimal.

code-documenter