note-to-blog

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill explicitly instructs Level 3 to read full note texts from the user's repo and append them to LLM prompts and to produce blog-ready drafts, which means any secrets embedded in notes (API keys, passwords, tokens) would be passed into the model and could be emitted verbatim — an exfiltration risk.