cite-check

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

• Third-party content exposure detected (high risk: 0.75). Outsider free text can enter the LLM context via the operating user’s drafts: cite-extract.ts reads markdown files from --drafts (user-selected files) and buildGroupPrompt() embeds the extracted claim/bodyContext/footnoteContext into the Gemini prompt in cite-check.ts → queryCitationFileSearch()/batch/generateContent.