external-content-sanitizer

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow ingests OUTSIDER-authored free text via the content argument (e.g., web-search/web-fetch results or cloned external repo files) into Step 4’s LLM analysis prompt as CONTENT TO ANALYZE: <the raw content>, which is indirect prompt-injection exposure (even though the sanitizer later removes/doesn’t echo matched text).