skill-scanner-hardened

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

• Third-party content exposure detected (high risk: 0.70). The Streamlit UI (streamlit_ui.py) explicitly accepts arbitrary uploaded files/ZIPs via file_uploader and pasted code via the text_area and then runs SkillScanner over that untrusted user-provided content to produce verdicts, so external/user-generated content is ingested and can influence the scanner's decisions.