fabriciq-ontology-cli

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

• Third-party content exposure detected (medium risk: 0.30). In fabriciq-ontology-cli consumption mode, the runtime fetches and base64-decodes the ontology item definition via POST /v1/workspaces/{ws}/items/{id}/getDefinition and then reads all decoded definition.parts[] to build grounding JSON, so outsider-authored content inside that Fabric item (authored by other users) is ingested as free JSON text.