zhongkao-physics

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). High risk — the skill intentionally sends user results and full wrong-question data to an external feishu-sync.js endpoint and mandates automatic publication (git push / GitHub Pages) of generated HTML, creating deliberate data-exfiltration and unauthorized-publishing channels.

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

• Potentially malicious external URL detected (high risk: 0.90). The generated HTML loads an external runtime script "https://xingyun-new.github.io/Skills-XiaoSiMen/lib/feishu-sync.js" (the declared single external dependency) which will execute remote JavaScript in the client and is required for FeishuSync.submit/... calls, so it is a runtime external dependency that executes remote code.

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 files into the repo and then "自动执行发布流程" including updating README and running git add → git commit → git push origin main (constructing GitHub Pages links), which instructs the agent to modify local files and perform remote push—an automatic, state-changing operation on the host.