Chinese ChatGPT Prompts: From Role Simulation to Verifiable Tasks
Purpose, mechanisms, getting started and boundaries
Source: PlexPt/awesome-chatgpt-prompts-zh · PlexPt
User Problem
Positioning and task fitawesome-chatgpt-prompts-zh collects Chinese role prompts illustrating task, tone, format and interaction constraints. It is a prompt library, not a tool installation. Distinguish simulated responses from real operations, then turn broad roles into tasks with inputs and acceptance criteria.
- Study format and interaction constraints in Chinese prompts
- Start translation, interview practice and writing tasks
- Adapt role templates into repeatable tasks
Product Mechanism
Capabilities and execution- Roles organize contextInterviewer and translator roles shape interaction and output. They can change response style without establishing professional credentials or real-world access. Source
- Output constraints add specificityExamples constrain output structure and turn-taking. Retain useful constraints while removing those that suppress necessary explanations or uncertainty. Source
- A simulated terminal is not a shellThe terminal prompt asks for text resembling terminal output; it does not connect to a real machine. Browser, database and interpreter roles likewise cannot establish execution. Source
- Machine-readable materialThe README points to prompts-zh.json for applications. Inspect fields, purpose and contents before importing templates rather than loading every role as system instructions. Source
Workflow
Start with a small task- Choose a bounded taskChoose translation, outlining or interview practice, then define actual inputs and expected results rather than copying only a job title.
- Replace the contextSpecify audience, language, evidence scope and facts to preserve, removing irrelevant example material.
- Add acceptance criteriaMark unknowns and check task-specific criteria: terminology for translation, role relevance for interview questions.
- Keep repeatable examplesTest different inputs and record failure conditions. Use observable tools for real search or execution, keeping their outputs distinct from generated prose.
The trial sequence and checks are editorial guidance. Entry commands follow official documentation; this project was not executed for the article. Source
Editorial prompt example
任务：把下面的英文发布说明翻译成中文。
约束：保留产品名、版本号、命令和事实；不补充原文没有的功能。
输出：译文，以及无法确定的术语。Complete the prerequisites above first. Examples contain no real credentials.
Limits and Boundaries
Constraints and selection| Boundary | Meaning | Suggested action |
|---|---|---|
| Professional-role credibility | The collection includes medical, legal and investment roles; role text does not provide professional review. Source | Do not rely on the role label as validation for consequential decisions. |
| Older examples and promotions | The README mixes promotion and older prompting patterns, including constraints that can hide explanation or simulate capabilities. Source | Select and adapt entries individually rather than treating the collection as one best practice. |
Scroll horizontally to view the full table
It is useful for learning prompt structure and inspiration. Production use needs domain inputs, explicit constraints and checkable output.
How to Evaluate Adoption
Trial scenario and verificationFor interview practice, retain one-question-at-a-time pacing and add seniority, subject scope and feedback criteria. Ask for specific omissions and improvements after each answer. The template then becomes an assessable exercise rather than generic role play.
Check scope, fact preservation, simulation-versus-execution distinction and performance on a different input.
Evidence: Pinned README · Current repository。
Source fetched (UTC): 2026-11-02T07:29:44.172Z；snapshot commit: a7085311b98a。The source date denotes the repository snapshot commit, not the project’s original launch.