fountain

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md instructs the agent to connect to Fountain via the Membrane CLI and to run/list actions (see "Connecting to Fountain" and "Running actions"), which will fetch and ingest user-generated Fountain documents/projects through a third-party service and thus could contain untrusted instructions that influence subsequent actions.

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 external Membrane CLI (installed via "npm install -g @membranehq/cli@latest", fetched from the npm registry e.g. https://registry.npmjs.org/@membranehq/cli) which is a required runtime dependency that downloads and executes remote code to provide actions used to drive the agent.