workbuddy-channel-setup

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill contains multiple deliberate patterns that enable credential capture and data exfiltration: it instructs saving/reading secrets (clipboard/state files), configures callbacks to an external server (codebuddy.cn) to receive messages/events, and uses injected eval (React fiber) in page context to bypass UI restrictions — together these enable sensitive-data collection and remote access via bot webhooks.

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

• Third-party content exposure detected (low risk: 0.10). It requires navigating a headed browser to third-party developer/console pages (e.g., open.feishu.cn / q.qq.com) where the automation can read and interact with page text, but no workflow step accepts outsider-authored free text via an attacker-controlled feed/queue.