skill-workflow-orchestrator

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

• Third-party content exposure detected (medium risk: 0.65). The workflow orchestrator parses and executes user-described multi-step chains that can include runtime-fetched public web content (e.g., via a “multi-search-engine”/“Fetch Data” sub-skill), which would ingest outsider-authored free text from those pages into the agent’s LLM context through the sub-skill outputs.