auto-extract

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt requires reading user messages, session logs and config files and then reporting/persisting specific message snippets and file contents (e.g., in AskUserQuestion descriptions and edits) with no instructions to redact secrets, so any API keys/passwords present would be extracted and output verbatim.