apply

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill explicitly instructs the agent to fetch SDK keys (via MCP or user paste) and write them into files like .env (and to perform file edits), which requires the LLM/process to handle and output secret values verbatim even though the doc asks not to echo them in chat.