project-guidelines-example

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt embeds real-looking secret values in the example environment variables (e.g., ANTHROPIC_API_KEY=sk-ant-..., SUPABASE_KEY=eyJ...), which can cause an LLM to output or copy those secrets verbatim into generated configs or commands, creating an exfiltration risk.