docs-check
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external documentation content to validate local files and suggest Pull Requests, creating a surface area for indirect prompt injection.
- Ingestion points: Markdown files downloaded from
https://code.claude.com(as specified inreferences/doc-pages.md). - Boundary markers: Documentation is passed to worker agents within the prompt context; no explicit delimiters or sanitization instructions are defined in the provided rules.
- Capability inventory: The skill can scan local directories, generate validation reports, and offer to create Pull Requests based on documentation content.
- Sanitization: No explicit sanitization or filtering of the fetched documentation is mentioned.
- [COMMAND_EXECUTION]: The skill executes a local shell script (
scripts/fetch-claude-docs.sh) to synchronize documentation from official sources. - [EXTERNAL_DOWNLOADS]: Fetches documentation pages from the official domain
code.claude.comfor use in validation tasks. This source is recognized as the official documentation for the platform.
Audit Metadata