optimize-md

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the model to read files and "preserve exact commands, configs, and code snippets" and to rewrite the file in place, which forces the LLM to read and output any secret values (API keys, tokens, passwords) present verbatim in the document, creating high exfiltration risk.