cusdis

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly directs the agent to connect to the third-party Cusdis service via the Membrane CLI and run actions that return Cusdis data (see "Connecting to Cusdis" and "Running actions"), which can include untrusted, user-generated comments that the agent would read and could influence subsequent behavior.

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

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