开发技能

Help a new contributor prepare the Chops development environment, generate its Xcode project, and build the app. The skill also explains the main code areas and common tasks, such as adding a tool source or changing skill parsing and the UI.

中文简介正在准备中。当前显示英语简介。

• Preparing a Mac for Chops development
• Generating and building the Xcode project
• Learning where app features are implemented

Use this skill to assess architecture, technology choices, delivery capacity, infrastructure, and technical debt in business terms. It helps compare build and buy options, surface costs and assumptions, and judge a technical plan before committing.

中文简介正在准备中。当前显示英语简介。

• Comparing whether to build or buy a capability
• Reviewing an architecture or technology choice
• Assessing delivery capacity and project risk

Use this skill to isolate repository work in a separate branch or worktree, keep it aligned with its base, and prepare it for review. It guides clean integration through merging, rebasing, or splitting work into smaller branches.

中文简介正在准备中。当前显示英语简介。

• Isolating a feature from current work
• Running parallel efforts in one repository
• Preparing a finished change for review

Design and review APIs around the concepts their users need. Shape resources, errors, pagination, and versioning, and assess whether a proposed change could break existing consumers so teams can plan compatibility and deprecation clearly.

中文简介正在准备中。当前显示英语简介。

• Design a new API interface
• Review an API before release
• Assess whether a change breaks consumers

Turn a specification or requirement into a step-by-step implementation plan with explicit changes, proof of completion, and review checkpoints. Then track execution in order, updating the plan when real conditions differ from expectations.

中文简介正在准备中。当前显示英语简介。

• Planning a change that spans several files
• Preparing a task for another agent or teammate
• Keeping a growing implementation within scope

Assess code and systems to separate deliberate technical debt from accumulated mess, outdated assumptions, or personal taste. Measure where debt costs delivery time or causes incidents, then prioritize incremental fixes and explain their business value to decision makers.

中文简介正在准备中。当前显示英语简介。

• Prioritizing technical debt by its ongoing cost
• Deciding whether to fix or accept an awkward system
• Explaining remediation value to non-engineers

Use this skill to verify that a change meets the original request before reporting it as complete. It guides agents to run the project’s real checks, read the output, look for side effects, and state clearly what remains unverified.

中文简介正在准备中。当前显示英语简介。

• Checking a fix before calling it done
• Reviewing project checks before a commit or pull request
• Reporting work with unverified parts

Clarify the outcome behind a request before implementation, then compare genuinely different approaches by cost, tradeoffs, and assumptions. It helps teams make a reasoned recommendation for a feature or behavior change, especially when the request is ambiguous.

中文简介正在准备中。当前显示英语简介。

• Clarifying an ambiguous feature request
• Comparing approaches before implementation
• Pressure-testing a design decision

Shape a system or major component around real constraints such as load, latency, data obligations, and team capacity. Compare architectural options, explain their costs and trade-offs, and record decisions so future teams understand the reasoning.

中文简介正在准备中。当前显示英语简介。

• Designing the structure of a new system
• Comparing architectural options and their trade-offs
• Reviewing where component boundaries should sit

Review code changes for correctness, downstream impact, security and data risks, tests, and design. It also helps assess feedback on your own work, decide which suggestions to implement, and explain a reasoned disagreement with evidence before a change is merged.

中文简介正在准备中。当前显示英语简介。

• Reviewing a diff or pull request before merging
• Checking a change for correctness and security risks
• Deciding how to address review comments

Use a red, green, refactor workflow: write a test that fails for the intended behavior, make the smallest change that passes, then clean up. Apply it to features, bug fixes, regression tests, and decisions about what behavior needs coverage.

中文简介正在准备中。当前显示英语简介。

• Starting feature implementation with a failing test
• Reproducing a bug before fixing it
• Changing difficult code with behavior checks

Investigate bugs, failing tests, and unexpected behavior by reproducing the issue, narrowing where it originates, and testing a specific hypothesis. Then verify that the proposed fix addresses the proven cause, rather than only hiding the symptom.

中文简介正在准备中。当前显示英语简介。

• Investigating a failing test
• Tracing an error after a code change
• Finding the cause of intermittent behavior

Design a repository’s multi-agent structure by assigning exclusive file areas, defining builder and reviewer roles, and checking ownership rules with a CI script. Use it to divide work, prevent overlapping edits, and decide whether another agent is needed.

中文简介正在准备中。当前显示英语简介。

• Dividing repository work among agents
• Defining agent roles and file ownership
• Auditing overlapping agent responsibilities

Use this skill to add Redis-based rate limiting to a TypeScript application. It covers SDK setup and basic usage, then points to guidance on algorithms, features, pricing, and traffic protection so you can choose and apply an appropriate limiter.

中文简介正在准备中。当前显示英语简介。

• Setting up a Redis rate limiter in a TypeScript app
• Choosing a rate-limiting algorithm
• Checking Redis pricing and operating costs

Visually inspect desktop frontend component changes by mounting them in a temporary Vite page. This lets you check the rendered UI in a browser without starting the Tauri app or a live model, then clean up the temporary harness files.

中文简介正在准备中。当前显示英语简介。

• Checking a desktop component's visual changes
• Previewing frontend work without launching Tauri
• Inspecting a component in a temporary Vite page

Read a text file from a given path so an agent can inspect configuration values, review code, or analyze data. It uses UTF-8 encoding and is intended for text files; binary files need separate handling.

中文简介正在准备中。当前显示英语简介。

• Checking values in a configuration file
• Reviewing the contents of a code file
• Reading text data for analysis

Search file contents across directories with a regular expression and get matching lines with their file paths and line numbers. Filter by filename, ignore case, limit results, or include nearby context. It uses Python and needs no external search tools.

中文简介正在准备中。当前显示英语简介。

• Finding function or class definitions across a project
• Locating TODO or FIXME markers, including mixed case
• Viewing matching lines with surrounding context

Modify an existing file by replacing an exact text match with new text, optionally changing every match. This focused editing method supports Windows and Unix line endings and is suited to small, targeted updates such as renaming a function or variable.

中文简介正在准备中。当前显示英语简介。

• Renaming a function in an existing file
• Replacing a variable name throughout a file
• Making a targeted text change while preserving other content

Retrieve detailed instructions for a specified MCP server, including its capabilities, tool usage guidance, examples, and best practices. Use the returned information to learn how that server works and how to apply its tools in a task.

中文简介正在准备中。当前显示英语简介。

• Learning what a particular MCP server can do
• Checking how to use a server's tools
• Finding server-specific examples and practices

Reload a skill that is already loaded after you edit its SKILL.md or scripts. The skill is unloaded, its instructions are parsed again, and it is registered with the updated changes. Use this to refresh a skill configuration during development.

中文简介正在准备中。当前显示英语简介。

• Applying edits to a skill's SKILL.md
• Refreshing a skill after updating scripts
• Re-registering an existing skill

Create interactive, multi-component web artifacts with React, TypeScript, Tailwind CSS, and shadcn/ui. The workflow initializes a project, supports state and routing, and bundles the finished work into one shareable HTML file for Claude conversations.

中文简介正在准备中。当前显示英语简介。

• Building an interactive multi-screen demo
• Creating a web artifact with state and routing
• Bundling a React project into one HTML file

Turn a specification or requirements for a multi-step development task into a detailed implementation plan. Break work into small actions and identify files, code, tests, documentation, and commands needed for each step.

中文简介正在准备中。当前显示英语简介。

• Planning implementation from a feature specification
• Breaking a large development task into small steps
• Listing files and tests needed before coding

Use this skill to guide feature work, bug fixes, and refactoring with test-driven development. It has the agent write a focused test, confirm that it fails for the expected reason, add the smallest implementation, and refactor while keeping tests green.

中文简介正在准备中。当前显示英语简介。

• Adding a feature with tests first
• Reproducing a bug before fixing it
• Refactoring code while checking behavior

Create Mermaid diagrams for workflows, system architecture, database models, schedules, and other common patterns. Choose a visual theme, render as SVG, PNG, or plain text ASCII, and generate a related set of diagrams in one batch.

中文简介正在准备中。当前显示英语简介。

• Show an approval flow as a diagram
• Map services and system architecture
• Visualize database relationships