Claude Skills by ssiumha

Manages parallel Claude instances via tmux for multi-worktree development. Use when spawning workers, checking status, collecting results, or coordinating parallel tasks.

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

Creates and manages custom Claude Code sub-agents. Use when automating workflows as dedicated agents, delegating recurring tasks, or modifying .claude/agents/ files.

Creates ast-grep patterns for structural code search. Use when finding functions/classes by structure, refactoring code, or when grep returns too many false positives.

Writes behavior specifications in Gherkin format. Use when defining features from user perspective or creating acceptance tests.

Sets up CI/CD workflows both locally (Justfile, lint, test) and remotely (GitHub Actions, GitLab CI). Use for test automation, pre-commit hooks, or deployment pipelines.

Python으로 Claude Agent SDK 에이전트를 구축할 때 사용. 커스텀 도구, 훅, 서브에이전트, MCP 연동 코드를 생성합니다.

Guides Claude Code setup and configuration. Use when setting up CLAUDE.md, rules, hooks, or skills structure for projects.

Provides CLI design guidelines and best practices. Use when developing command-line tools, designing flags, help text, or error messages.

Measures code structure metrics (coupling, cohesion, complexity, call graph). Use when analyzing code quality quantitatively, checking architectural health, or reviewing structural changes.

Python type hints and typing review checklist. Injected into code-reviewer agent for Python projects.

TypeScript type safety review checklist. Injected into code-reviewer agent for TypeScript/TSX projects.

Creates slash commands for repeated workflows. Use when defining batch processing, pipelines, or checklist-based automation.

Provides DDD layered architecture patterns. Use when designing domain-centric systems or structuring complex business logic.

Automates deployments and infrastructure management. Use for PM2, Helmfile/K8s, Ansible playbooks, database config, monitoring setup, or multi-server automation.

Creates Docker configurations and compose files. Use when containerizing apps, setting up Docker dev environment, writing Dockerfile/compose.yaml, or building multi-stage images.

전자금융감독규정 인증 준비 지원. 조항 조회, 체크리스트 생성, 이행 계획 수립.

Provides fzf command patterns and configurations. Use when writing fzf commands, interactive filters, or pipeline compositions.

Resolves Git conflicts (rebase, merge, cherry-pick, stash). Use when encountering merge conflicts during any Git operation.

Manages git worktrees for parallel development. Use when creating feature branches, switching between tasks, managing multiple work streams, or cleaning up after PR merge.

Provides spec writing guidelines with 6 core areas and boundary system. Use when writing SPEC.md, defining requirements, or creating project specifications.

Configures Claude Code hooks for projects and skills. Use when setting up lint, test, or format hooks in .claude/ directory, or applying hook recipes to skills via SKILL.md frontmatter.

ISMS-P 인증 준비 지원. 항목 조회, 체크리스트 생성, 이행 계획 수립.

Manages project knowledge base and TODO tracking. Use when recording decisions (ADR), creating TODOs, updating knowledge docs, checking "what to do", or organizing project documentation.

Audits lint errors, type checks, dead code, and commit hygiene. Use when checking project health, running lint/type checks, or finding unused code.

Adds docstrings, comments, and WHY/DECISION tags to code. Use when documenting functions/classes with language-specific standards (JSDoc, PEP 257, GoDoc).

Generates mise.toml project configuration. Use when setting up project tools, environment variables, or task automation with mise.

Initializes multi-repo workspace CLAUDE.md. Use when managing multiple repositories as a single project.

Creates dependency-based task execution plans with parallel orchestration. Use when 3+ tasks with dependencies, unclear execution order, or parallel work identification needed.

Generates project context (code structure + architecture intent). Use when starting sessions, after major refactoring, or when delegating complex work to agents.

Refines user prompts through structured dialogue to produce executable prompts with delegation plans. Use when user requests are vague or multi-file, or manually via /prompt-review or /refine.

Searches local markdown notes and documents using qmd CLI. Use when searching notes, querying documents, managing collections, or retrieving document content.

방향 수정 신호 감지 및 세션 전체 회고. "아니/잠깐/근데" 감지 시 자동 트리거, /reflect 또는 "회고해줘"로 세션 회고.

Reviews code for security vulnerabilities. Use when implementing auth, handling user input, adding APIs, or before deployment. Checks OWASP Top 10, credentials, injection.

Creates and manages Claude Code rules files. Use when creating .claude/rules/ files or applying conditional rules to specific files/situations.

Creates and manages Claude Code skills. Use when creating new skills, updating existing skills, or converting repeated prompts into reusable skill packages.

프로젝트별 외부 skill 설치 및 관리

Provides Spring Boot API development standards. Use when implementing response formats, logging, i18n, or exception handling patterns.

Applies TDD workflow to development. Use when implementing features, fixing bugs, or refactoring with test-first approach. Detects AI anti-patterns.

Sets up Claude Code agent teams with role-based composition and PM task management. Use when creating dev teams, defining team roles, setting up task tracking with backlog.md, or organizing multi-agent collaboration. Do NOT use for single sub-agent creation (use agent-creator instead).

Guides test writing and review. Use when writing new tests, reviewing existing tests, or diagnosing mock overuse (see Workflow 2).

Performs small structural code cleanups (tidyings). Use when preparing code changes, removing dead code, reducing nesting, or cleaning up before feature work.

Automates tmux pane/window orchestration for Claude instances. Use when running dev servers in separate panes, monitoring logs across panes, checking process status, sending commands to other panes, or when background tasks need persistent terminal output instead of run_in_background.

UI 디자인 기본 원칙 가이드. Use when creating UI components, writing frontend code, reviewing designs, or avoiding AI-looking interfaces.

KISA/금융보안원 기술적 취약점 점검항목 제공. 서버(Unix/Windows), 웹 모의해킹 점검 시 사용.