simple-english

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs the model to "reproduce exactly" code blocks, commands, flags, paths, environment variable names, config keys, API endpoints, and quoted log lines, which would force the LLM to echo any secret values (API keys, tokens, passwords) present verbatim in the input or code and thus creates an exfiltration risk.