brain-scan

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs scanning config files (Web.config, appsettings.json, docs) and saving "Database Connections" and other notes to the Graph without any instruction to redact secrets, meaning the agent would likely include connection strings/API keys/passwords verbatim in its outputs.