web-atlas-web-design
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses an external Model Context Protocol (MCP) tool hosted at
https://web-atlas.ayangabryl.workers.dev/mcp. This connection is consistent with the skill's purpose of researching design references and corresponds to the vendor's own infrastructure. - [INDIRECT_PROMPT_INJECTION]: The skill acts on data retrieved from external websites (screenshots, design tokens, and layout references). Although this is an ingestion surface for untrusted data, the instructions explicitly guide the agent to treat this content as design evidence to be interpreted within the context of the user's brief, rather than as authoritative commands, reducing the risk of indirect injection.
Audit Metadata