google-workspace

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The code deliberately routes OAuth flows and token refreshes through a third‑party cloud function (default https://google-workspace-extension.geminicli.com), including POSTing refresh tokens and using that endpoint as the OAuth redirect, which constitutes intentional transmission of sensitive credentials to an external service and is a high-risk data‑exfiltration/credential‑theft pattern.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's exec runner and examples (scripts/workspace.js and SKILL.md) explicitly call Google Workspace APIs (e.g., workspace.call('drive', ...), docs, gmail) to fetch user-generated third-party content (Drive/Docs/Gmail) that the agent is expected to read and could materially influence subsequent tool use or decisions.