The Agent Skills Directory

[SAFE]: The skill is designed to facilitate the generation of API documentation from a codebase. It describes a workflow for analyzing routes, methods, and schemas to produce markdown documentation, OpenAPI specifications, and Postman collections.
[CREDENTIALS_UNSAFE]: The examples provided in the skill use well-recognized placeholders for sensitive information, such as YOUR_TOKEN, your-password, and refresh_token_here. A partial JWT string (eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9...) is used in an example, but it represents only a generic header ({"alg":"HS256","typ":"JWT"}) and is clearly marked as illustrative content.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is instructed to 'examine your API codebase' (Step 1), which involves processing untrusted source code and comments. However, since the skill's capabilities are limited to generating descriptive text and structured documentation (markdown, JSON, YAML) without executing the code or performing network operations, the risk is negligible.

api-documentation-generator