website-to-hyperframes

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

• Third-party content exposure detected (medium risk: 0.30). Step 1’s capture runtime path runs npx hyperframes capture <URL> and then reads LLM-ingestible extracted files like extracted/visible-text.txt, extracted/tokens.json, and extracted/asset-descriptions.md derived from the outsider-provided website URL.