security-testing

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

• Third-party content exposure detected (medium risk: 0.65). The orchestrator’s runtime workflow explicitly invokes multiple specialist skills (e.g., recon-and-osint, active-pentest) that commonly fetch public web content or process user-supplied targets, which can introduce outsider-authored free text into the LLM context via those skills’ ingestion of fetched pages/messages.