narrative-voice-coach

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

• Third-party content exposure detected (medium risk: 0.30). [The runtime workflow takes user-supplied free text as input (e.g., voice match [text], voice consistency [chapter], and “Read them before you start writing” from YOUR work) and then the LLM ingests that text to analyze/generate coaching outputs; this is outsider-authored content submitted by the user.]