webmcp-builder

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a developer guide and scaffolding tool for WebMCP development.
  • [PROMPT_INJECTION]: The skill addresses indirect prompt injection surfaces where tools ingest untrusted web data.
  • Ingestion points: Fetch results from external APIs and user-submitted forms (reference/security.md).
  • Boundary markers: Recommends the use of [USER CONTENT START] and [USER CONTENT END] delimiters.
  • Capability inventory: Tools can perform network requests and manipulate page elements.
  • Sanitization: Recommends text truncation and structured response schemas.
  • [EXTERNAL_DOWNLOADS]: Mentions the official GoogleChromeLabs repository for testing, which is a well-known trusted source.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 05:08 PM