book-chunk

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text is ingested from the runtime-walked <book-slug>-ch*.md chapter files (Step 1/2), whose body text is then chunked and embedded into chunks.jsonl and thus fed into the LLM for the per-chunk contextual prefix generation (Step 4).