wormhole

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can be ingested into the agent’s LLM context via the runtime RESPONSE=$(maw hey $RESOLVED "$MESSAGE" ...) path, where the remote oracle’s returned output (authored by a non-operating party) is printed/returned and thus becomes readable text for the agent to process.