daily-cowork-backup

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly reads and prints cfg["apiKey"] and cfg["url"] in Step 1 and instructs substituting those real values into the generated upload script, which requires the agent to expose secret values verbatim in its output or generated files.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill intentionally collects Cowork VM files (including a cowork-gdrive-config.json credential file), stages them to the host, and uploads them to an external Google Apps Script endpoint while instructing deletion of local screenshots/staging—behavior consistent with deliberate data exfiltration and credential theft (with steps to remove traces).