fumadocs
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a documentation assistant for the Fumadocs framework. It provides specific templates for a research sub-agent to fetch information from the official domain
fumadocs.dev. - [INDIRECT_PROMPT_INJECTION]: The skill establishes an external data ingestion surface by fetching documentation from
https://www.fumadocs.dev/. This is identified as a low-risk surface for indirect prompt injection. Evidence Chain: (1) Ingestion point is the official Fumadocs documentation site; (2) Boundary markers are implemented through structured prompt headers for the sub-agent; (3) The research sub-agent is explicitly configured withreadonly: true, preventing write operations or state changes; (4) The retrieved content is informational and intended for developer guidance.
Audit Metadata