discipline

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill requires the agent to quote user-provided wording verbatim ("quote their wording" and "Anything quoted or labeled 'verbatim' MUST be character-for-character"), which forces the LLM to reproduce exact user strings (and thus any secrets the user supplies) in its output, creating an exfiltration risk.