siteleaf

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to fetch and interact with arbitrary Siteleaf site content via Membrane (e.g., using "membrane action list"/"membrane action run" and "membrane request" to call the Siteleaf API), which will ingest untrusted, user-generated third‑party webpages/posts/assets that the agent is expected to read and could materially influence subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill requires installing and running the Membrane CLI (npm install -g @membranehq/cli@latest and npx @membranehq/cli@latest), which fetches and executes remote code from the npm registry (e.g. https://www.npmjs.com/package/@membranehq/cli or https://registry.npmjs.org/@membranehq/cli) at runtime, making it a required external dependency that executes remote code.