webmcp-tool-annotations
Pass
Audited by Gen Agent Trust Hub on Mar 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: Fetches technical specifications from the official Web Machine Learning community documentation site (webmachinelearning.github.io) for standard protocol reference.\n- [SAFE]: Provides instructional content focused on enhancing security via tool safety hints and browser permission prompts.\n- [SAFE]: Explicitly warns that incorrect annotations are a security risk, noting that marking a destructive tool as read-only could bypass safety filters, and emphasizes that annotations are hints requiring backend enforcement.
Audit Metadata