effect-command-executor

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt includes examples that embed environment variables (e.g., env: { API_KEY: 'xyz' }) and shows building commands/options programmatically, which can lead an LLM to output secret values verbatim in code/commands even though it doesn't explicitly require soliciting keys from users.