appwrite

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md workflow explicitly instructs the agent to run actions such as "list-documents" and "get-file" and to proxy arbitrary requests to an Appwrite API via Membrane, which will fetch and expose user-generated/untrusted Appwrite documents and files that the agent is expected to read and that could materially influence subsequent actions.