nlm-skill

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs extracting cookies from Chrome DevTools and passing them as cookies="<cookie_header>" to mcp__gemini-notebook-mcp__save_auth_tokens(), which requires embedding sensitive auth cookie/header values verbatim in generated commands/calls.

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

• Third-party content exposure detected (medium risk: 0.30). Workflow path reads outsider-authored free text from user-accessible NotebookLM sources—either pasted text via nlm source add ... --text/MCP source_add(source_type="text", ...) or web/drive content first added by the user and then ingested for nlm notebook query/studio_create.