webmcp

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (low risk: 0.10). The workflow opens user-provided or aliased URLs in a browser and executes WebMCP tool calls on arbitrary web pages, which is a low-risk user-selected URL fetch.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (medium risk: 0.30). The installation command in README.md uses npx to fetch and execute a package from npm (@agent-native/skills@latest), representing an unverifiable runtime external dependency under common/reputable unofficial publisher provenance.