writing-plans

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). This prompt's rules ("Every step shows actual content", "No placeholders", "Exact commands") effectively force the author to embed concrete values in code/commands, which can lead an LLM to include API keys, tokens, or passwords verbatim rather than using safe env vars or CLI-auth — creating an exfiltration risk.