remember

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's workflow takes the user's input verbatim, appends it to MEMORY.md, and echoes a confirmation including the exact entry, so if a user supplies credentials or tokens the agent would write and output those secrets (despite a warning not to); thus it enables secret exfiltration.