geofeed-tuner

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

• Third-party content exposure detected (high risk: 1.00). The skill's required workflow explicitly downloads and ingests CSV input from remote URLs in Phase 2 (saving to ./run/data/) and then uses those parsed entries (and external MCP responses from https://mcp.fastah.ai/mcp in Phase 4) to drive validation, tuning, and next actions, so arbitrary public web content and third‑party responses can materially influence the agent's behavior.