azure-openai-to-responses

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow scans a user-provided/local code directory (via python skills/.../detect_legacy.py .) and reads file contents directly with Path.read_text() for regex matching, so outsider-authored free text in those files is ingested.