The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes component names and documentation titles from external URLs and the local codebase, which could contain malicious instructions designed to influence the agent's output.\n
Ingestion points: Reads Storybook index.json from remote URLs, crawls HTML for page titles, and parses local component and CSF files.\n
Boundary markers: There are no defined delimiters or instructions to treat ingested content as untrusted data.\n
Capability inventory: The skill is configured with Write access and several Bash tools for file and history inspection.\n
Sanitization: Metadata gathered from external and local sources is not sanitized before being included in the documentation audit report.\n- [EXTERNAL_DOWNLOADS]: The skill connects to remote Storybook and documentation platform endpoints to retrieve inventory data.\n
Evidence: The configuration allows for fetching index.json from a Storybook URL and accessing Zeroheight or Supernova APIs.

docs-coverage