개발 스킬

Delegate independent implementation tasks or investigations to separate agents, then review their results at checkpoints. The workflow supports sequential work for related tasks and parallel work for unrelated problems, with verification and integration steps to catch issues.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Implementing a multi-step plan with separate tasks
• Investigating unrelated failures in parallel
• Reviewing agent changes before continuing

Prioritize changed repository files for human review by combining agent assessments of the change story, impact, failure risks, and expectations. It produces a focused file list, key facts, a random sample of changed files, and declarative files for review.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Prioritizing files in a code review
• Reviewing staged and unstaged changes
• Checking a branch diff or a specific commit

Guides you through setting up the Serena MCP server for a chosen project or user scope. It helps identify your client, check prerequisites, configure the connection, initialize project indexing, and verify that Serena tools are available.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Setting up Serena for a project or user account
• Configuring Serena for Claude Code, Cursor, or another MCP client
• Initializing project indexing and checking the connection

Set up Context7 MCP and add relevant documentation library IDs to the appropriate CLAUDE.md file. Choose project-shared, project-local, or user-wide configuration, and identify technologies from the supplied input or project structure when needed.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Set up Context7 MCP for a project
• Choose shared, local, or global configuration
• Add documentation library IDs to CLAUDE.md

Trace a problem from its visible symptoms to underlying technical or process causes by asking “why” in repeated steps. The skill helps validate causes by working backward, explore multiple branches, and suggest solutions that address the root cause.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Investigating recurring production errors
• Finding causes of flaky CI/CD runs
• Tracing delays in feature deployment

Set up Codemap CLI for a project or user account, install it, verify it, and add configuration guidance and hooks. Codemap presents codebase structure, dependencies, and changes to help agents navigate a project.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Set up Codemap for a shared project
• Install Codemap for personal use
• Add hooks that show codebase context at session start

Use Git notes to attach review status, test results, audit details, or other metadata to commits without changing their hashes or rewriting history. This skill covers adding, viewing, organizing, sharing, merging, and preserving notes across rebases.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Recording code review status on commits
• Adding test results to specific commits
• Sharing notes with a Git remote

Analyze a project’s tools and settings, suggest suitable Claude Code hooks, then configure and validate them. Hooks are commands triggered by events such as file edits or before tool use; examples include formatting code, running checks, and blocking sensitive edits.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Suggesting hooks for a project’s tools
• Running checks after code edits
• Blocking changes to protected files

Create a GitHub pull request with GitHub CLI using the project’s required title format and pull request template. Check for uncommitted changes first, create the request as a draft when work is in progress, and use English for its title and description.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Creating a draft pull request from committed work
• Preparing a PR description that follows the project template
• Checking PR status or marking a draft ready for review

Choose a Kaizen analysis method for a codebase, workflow, or process. The skill guides a Gemba Walk to inspect actual code, Value Stream Mapping to find delays, or Muda analysis to identify waste, then presents actionable findings.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Exploring how an unfamiliar feature actually works
• Finding delays in a deployment or review workflow
• Identifying duplication, technical debt, or over-engineering

Create Claude slash commands through an interactive process. Clarify the command’s purpose, choose a category and project or user location, then draft its structure, supporting resources, and documentation using established patterns.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Designing a command for feature planning
• Creating a reusable implementation command
• Adding a project-specific analysis command

Review only unresolved pull request threads and check each comment against the current files, including uncommitted local changes. Resolve threads only when every requirement is met or the concern is obsolete, and report which threads remain open with reasons.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Checking whether requested code changes satisfy review comments
• Resolving review threads after fixes are present locally
• Leaving partially addressed feedback open

Work on several Git branches in separate directories using Git worktrees. The skill covers creating and cleaning up worktrees, reviewing pull requests without interrupting current work, running isolated tests, and comparing or selectively combining changes.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing a pull request while feature work continues
• Testing changes in an isolated directory
• Comparing implementations across branches

Compare recent Git changes with the project’s FPF knowledge base to find context drift, evidence tied to changed files, and decisions that may need review. The skill reports findings and updates the baseline commit for the next audit.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Checking the knowledge base before starting new work
• Finding evidence affected by recent code changes
• Reviewing project decisions after configuration changes

Create or update agent skills by testing how agents handle realistic scenarios before writing instructions. The guide adapts the RED-GREEN-REFACTOR cycle from test-driven development: observe failures, write targeted guidance, and test again to close gaps.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Drafting a new agent skill
• Updating existing skill instructions
• Checking whether a skill works in practice

Review completed work from requirements, architecture, and code quality perspectives. Specialized judges assess it independently, verify their findings, and compare disagreements before producing an evidence-based report with prioritized issues and recommendations. The review does not change files.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing recent implementation work
• Checking changes against the original request
• Getting prioritized code improvement suggestions

Add review feedback directly to specific changed lines in a GitHub pull request. The skill explains how to post one comment or group several comments into a review, using an available inline-comment tool or GitHub CLI (`gh`).

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Comment on a specific changed line
• Group several code comments into one review
• Choose the correct diff line and side

Review staged changes and create concise commits using emoji paired with Conventional Commit types such as feat, fix, and docs. The workflow can run project checks, suggest splitting unrelated changes, and follow a branch naming pattern when starting from the main branch.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Writing a commit message that matches code changes
• Splitting unrelated changes into separate commits
• Running pre-commit checks before committing

Delegate a task to an implementation agent, have a separate agent set tailored evaluation criteria and judge the result, then retry with specific feedback when needed. The process allows up to three retries before escalating unresolved issues.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Implementing a feature with independent quality checks
• Refactoring code with feedback driven revisions
• Completing a task that needs a defined evaluation rubric

Review staged and unstaged local changes, including untracked files, before committing. You can focus the review on areas such as security or performance, set a minimum issue impact, and receive actionable findings in Markdown or JSON.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Reviewing local changes before a commit
• Focusing a review on security or performance
• Getting filtered review findings as JSON

Load unresolved review threads from a GitHub pull request and turn them into focused Markdown task files under `.specs/comments/`. Group related feedback without duplicating work, preserve actionable suggestions, and leave GitHub comments and PRs unchanged.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Turning unresolved PR feedback into implementation tasks
• Grouping review comments for parallel work
• Keeping resolved threads out of a review task list

Applies an incremental approach to coding, refactoring, system design, and workflow changes. It encourages error prevention through types and early validation, consistency with existing project patterns, and solutions sized to current needs instead of speculative complexity.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Making a small, focused improvement during implementation
• Designing validation and error handling to catch problems early
• Refactoring while keeping existing project conventions

Turn recurring agent mistakes or behavior gaps into focused `.claude/rules` files. The guidance uses incorrect and correct examples to make expectations clear, and helps distinguish always-on rules from task-specific skills or broad project notes.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Documenting a mistake an agent keeps repeating
• Adding a project-wide behavior constraint
• Scoping guidance to particular file paths

Build MCP servers that let LLMs work with external APIs and services. The guide covers workflow-focused tool design, implementation in Python with FastMCP or Node/TypeScript with the MCP SDK, error handling, review, and evaluations using realistic questions.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Design tools for a service integration workflow
• Implement an MCP server in Python or TypeScript
• Plan validation, pagination, and error handling