writing-style

Warn

Audited by Snyk on Feb 19, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 0.90). The README's "Step 1: Export Your Real Writing" and "Step 2: Ask AI to Find Your Patterns" explicitly instruct users to copy/paste messages from external, user-generated sources (e.g., Twitter/X posts, Gmail/WhatsApp/Slack exports) into Claude/ChatGPT for analysis, so the agent ingests untrusted third-party content that directly determines style rules and thus can materially influence its subsequent outputs.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 19, 2026, 01:08 PM