scrivener-extract

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

• Third-party content exposure detected (high risk: 0.75). The skill reads outsider-authored free text from the user’s .scriv project files at runtime by parsing each item’s content.rtf and converting it to LLM-ingestible prose (JSON content), and that RTF body text is not authored by the operating user/organization in general.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)