zhongkao-english

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). High-risk: the skill intentionally forces automatic publishing (git add/commit/push) and includes a remote script that submits detailed user/practice data to an external Feishu-sync endpoint — behavior consistent with deliberate data exfiltration and repository modification.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The HTML template embeds and calls an external script that will be fetched and executed at page runtime (FeishuSync), specifically: https://xingyun-new.github.io/Skills-XiaoSiMen/lib/feishu-sync.js, which provides the FeishuSync.submit API used in the page.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill mandates writing HTML files into the project, then forcibly executing an automated GitHub Pages publish flow including reading a publish script and running git add/commit/push (modifying repo files and performing network pushes), which instructs the agent to change the machine/repository state and perform remote operations.