verify-llm-artifacts

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill mandates echoing the parsed findings and reading/reporting file contents from .beagle/llm-artifacts-review.json (and cited files) verbatim before adjudication, so any API keys, tokens, passwords, or other secrets present in those findings/files would be output directly, creating an exfiltration risk.