creating-docs-diagrams
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Employs 'rg' (ripgrep) to locate Mermaid blocks and 'Playwright' to verify diagram rendering and viewport responsiveness.
- [INDIRECT_PROMPT_INJECTION]: The skill processes diagram definitions from markdown files, creating an attack surface for indirect prompt injection. (1) Ingestion points: Reads .mdx files in 'website/src/content/docs/'. (2) Boundary markers: Targets content specific to 'mermaid' code blocks. (3) Capability inventory: Includes file system searching, file modification, and browser execution via Playwright. (4) Sanitization: Enforces a strict visual contract that prohibits 'foreignObject', scripts, and external assets in the resulting SVG output to mitigate execution risks.
Audit Metadata