proapis

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs the agent to fetch and run actions against Proapis via Membrane (e.g., "membrane request CONNECTION_ID /path/to/endpoint" and "membrane action run ..."), which will ingest external Proapis data (user-generated/untrusted records like tasks/projects) that the agent is expected to read and act on, allowing third-party content to influence subsequent decisions.