The Agent Skills Directory

[SAFE]: The skill's primary purpose is to automate the creation and maintenance of component documentation in the Sentry repository. It defines a structured workflow for gathering context and generating MDX files.
[COMMAND_EXECUTION]: The skill uses tools like Grep, Read, and Glob to analyze existing source code and component usage. This is standard behavior for an agent designed to document an existing codebase.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data by reading existing project files and taking user input for documentation descriptions. However, it lacks dangerous capabilities like arbitrary network operations or high-privilege command execution, and the data is used primarily to generate static markdown content.
[EXTERNAL_DOWNLOADS]: The skill references resources from the official Sentry GitHub repository and well-known documentation sites (W3C, WAI-ARIA). These references are used for metadata and link generation in the documentation, which is a standard and safe practice.
[DYNAMIC_EXECUTION]: The skill uses the !!type-loader! syntax, which is a standard Webpack loader pattern used in the Sentry project to extract TypeScript definitions for documentation. This is a build-time configuration rather than a runtime execution risk.

react-component-documentation