testing

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly asks the user to paste a preview URL containing a short-lived JWT and instructs the agent to capture/use that exact tokenized URL (and store it in preview-token.local), which requires the LLM to accept and embed the secret token verbatim in navigation/commands/files — an explicit secret-handling/exfiltration risk.