ptz-docs

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

• Third-party content exposure detected (high risk: 0.85). At runtime the skill fetches and text-extracts HTML from outsider-authored public web pages on onvif.org (e.g., onvif-profiles, onvif-profile-s) via urllib.request.urlopen in fetch_url() and TextExtractor, then returns that extracted prose in search/section/fetch outputs that the agent can place into the LLM context.