icp-deep-scanner

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests outsider-authored free text from sources like public web reviews/pages via a WebFetch/customer-review-aggregator path (e.g., G2/Capterra/Trustpilot/App Store review body text), which is then summarized and written into icp-profile.md/personas/*.md as “verbatim language (scrubbed quotes)”.