wiki-from-research

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

• Third-party content exposure detected (high risk: 0.90). The skill orchestrates cogni-research and then runs "wiki-ingest --discover research:" which materialises per-sub-question files and explicitly lists "contexts, sources" and asks "Ingest these N sources?" (Workflow Steps 1 and 3), showing it consumes untrusted external research sources that the agent reads and uses to decide ingests and subsequent actions.