deployment

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

• Third-party content exposure detected (high risk: 0.70). The required runtime workflow includes checking human-interventions/active/ for files tagged phase: 09-deployment/all, and those files are free-text authored by humans outside the agent, which can be ingested into the LLM context via the “check human-interventions/active/” step.