hive-token-diligence

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow calls search_tools / invoke_api_endpoint for token metadata and risk, and those runtime tool responses (e.g., scraped web/provider content) would be ingested as free text into the agent context, creating an indirect prompt-injection surface from outside-authored data.