Claude Code / Use

Claude Code Prompt Dictionary

Turning high-frequency tasks into reusable prompts can reduce context waste and improve delivery consistency.

How to use the prompt dictionary

The prompt database does not allow you to copy a universal spell, but fixes the input structure of high-frequency tasks. Good prompts include goals, context, constraints, output format, and validation methods.

Learners can use the official prompt vocabulary as a template source, and then rewrite it based on their own repository commands, technology stack, design constraints and acceptance criteria.

Commonly used templates

Write each type of task into a clear engineering request.

Explain the code base

Specify the modules you care about and let Claude output the call chain, key files and risk points.

Fix bug

Provide errors, recurrences and expectations, requiring you to first locate the root cause and then make minimal repairs.

write tests

Describe the framework, boundary scenarios, whether mocking is allowed, and what commands to run.

review diff

It is required to list the real risks in order of severity and provide executable repair suggestions.

copyable structure

When using a prompt bank, it is more important to preserve the structure than the original sentence.

Goal: Fix the occasional 401 after the user refreshes the token.
Scope: src/auth, middleware, related tests.
Restrictions: Do not change payment, do not change database schema.
Process: Explain the reason first, and then submit the minimum diff.
Verification: Run auth tests and pnpm build.
Output: change files, verification results, remaining risks.

Complete usage points

Supplement the core concepts, operation sequences, permission boundaries and verification requirements that are easily compressed and missed in official documents, making it easier for English readers to learn completely by page.

How to use the prompt dictionary

The official prompt thesaurus is suitable as a structural template, rather than copying the original sentence. High-quality prompts include goals, context, constraints, output formats, and verification methods. Rewriting the template according to the actual commands of the repository will have a more stable effect than general prompts.

It is recommended to save high-frequency tips as team commands or skills: explaining the code base, fixing bugs, writing tests, reviewing diffs, generating migration plans, updating documentation, investigating performance issues, handling CI failures.

For complex tasks, use the "Don't change the file, investigate and plan first" tip first. After confirming the plan, enter the editing stage. This prevents Claude from making a premature move.

Study Checklist

Put the content on this page into real tasks and use the five dimensions of entry, context, permissions, verification and team rules to check whether you have truly mastered it.

Study Checklist

After reading this page, do not just remember the concept name. You should be able to place "Claude Code Prompt Dictionary" back into a real Claude Code engineering workflow: where the task starts, what context the system loads, which actions need approval, how the result is verified, and how to roll back when it fails.

If this is a portal or platform page, specifically confirm what contexts this portal can access: local files, cloud repositories, browser logins, team messages, external tools, and whether these contexts are sufficient to complete the verification.

Be able to describe in your own words the specific problem this page solves, rather than just reciting the title.
Able to write a minimal example task with goals, scope, prohibitions, and acceptance criteria.
Be able to determine which information should be put into the current prompt and which should be captured as project rules or configurations.
Be able to explain which long-term rules should go into CLAUDE.md, and which runtime behavior should be handled by settings, permissions, Hooks, Skills and MCP.
Ability to check diffs, command output, test results, screenshots or PR notes after a task is completed instead of just trusting the natural language summary.

If this page is used for team training, ask learners to complete a small task with Claude Code: read and explain first, submit a plan, make the smallest useful change, and close with real verification commands plus human diff review.