circle

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly shows fetching user-generated content from Circle (e.g., "List Posts", "List Comments", "Get Post") via Membrane CLI commands and proxy requests (membrane action run / membrane request), which would cause the agent to ingest and act on untrusted third‑party community content from Circle.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill requires installing and running the Membrane CLI via "npm install -g @membranehq/cli" (which fetches and executes remote code from the npm registry, e.g. https://www.npmjs.com/package/@membranehq/cli) and that external code is a required runtime dependency.