AI Agent Hub
Back to plugins
🖥️

dsh-civil-exam

Client Updated 2026.08.20

Run the following command in DeepSeek Harness:

dsh plugin install uckkk/dsh-civil-exam

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install uckkk/dsh-civil-exam inside DeepSeek Harness to install this plugin, source available at https://github.com/uckkk/dsh-civil-exam under the MIT license.

About this plugin

The administrative aptitude test (Xingce) is the most time-pressured section of China's civil service exam. Candidates often rely on a vague sense of how many minutes to spend per question, and that hesitation compounds under exam conditions. dsh-civil-exam replaces that gut-feel pacing with a set of precise, reproducible calculations so you can plan your question order and time budget with confidence before you ever sit down for the real test.

The plugin exposes a single tool, civil_exam, which takes parameters such as total question count, time limit, and per-item time allocation, then runs the computation entirely inside a pure Node.js runtime and returns a structured, detailed result. There are no network calls, no external services, and no heavy dependencies, so it works offline with near-instant response times. Once installed, you simply invoke the registered tool within a session and read the breakdown it returns.

If you are preparing for the Xingce and want a data-driven, repeatable way to calibrate your pacing rather than relying on trial-and-error from past mock exams, dsh-civil-exam is a lightweight, zero-dependency utility released under the permissive MIT license that you can pick up and use right away.

Use Cases

  • Calculate the optimal question order before the exam based on item count and type distribution
  • Analyze whether per-type time spending was reasonable after a mock exam
  • Build a personalized time budget for different difficulty combinations

Best For

  • Civil service exam candidates who need a quantified pacing strategy
  • Self-studying test-takers who prefer data over intuition for question ordering
  • Users who favor lightweight, offline tools with zero external dependencies