prompt-builder

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

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow repeatedly dispatches Researcher Subagent and Prompt Updater during the research/modify phases, which (by design) can ingest outsider-authored free text from runtime research sources (e.g., public web content or other external documents) into the agent context via the subagent’s research/update steps.