bmad-deep-recon

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

• Third-party content exposure detected (medium risk: 0.30). In Run mode, the workflow performs a web fan-out/acquisition loop and writes extracted digests derived from outsider-authored web content into {doc_workspace}/digests/, which the lead and verifier subagents then read (via the research firewall) to produce citations and claims.