digio

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md explicitly instructs the agent to fetch and run actions against the Digio API via Membrane (e.g., "membrane action run ..." and "membrane request CONNECTION_ID /path/to/endpoint"), which will return third-party file/content and links from Digio that the agent is expected to read and act on, exposing it to untrusted user-provided content.

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

• Potentially malicious external URL detected (high risk: 0.70). Flagged because the skill runs "npx @membranehq/cli@latest" (fetching and executing the package from the npm registry, e.g. https://www.npmjs.com/package/@membranehq/cli) at runtime as a required dependency, which results in remote code being fetched and executed locally.