By sakuga-cmd
18 specialized agents for full-stack development with unique Russian-speaking personalities
Use for writing tests (TDD), finding edge cases, analyzing test failures, test coverage improvement. Do NOT call for writing production code (that's Timur/Oleg) or deployment (that's Kate).
Use for accessibility audits, WCAG compliance checks, ARIA verification, keyboard navigation, color contrast. Do NOT call for visual design, backend work, or performance optimization.
Use for security audits, OWASP checks, XSS/CSRF/injection detection, access policy review, secrets audit. Do NOT call for feature development, UI work, or performance optimization.
Use for web research, finding documentation, comparing approaches, searching npm packages, reading external articles. Do NOT call for codebase exploration (that's Mark) or writing code.
Use for UI component design, visual design, design system work, animations, responsive layouts. Do NOT call for business logic (that's Timur/Oleg) or backend integration (that's Oleg).
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
18 specialized agents with unique personalities and nationalities for full-stack development. An international dev team — each agent has a distinct character, cultural flavor, clear responsibilities, and knows when to call teammates.
# Add marketplace
/plugin marketplace add sakuga-cmd/claude-code-agents
# Install plugin
/plugin install claude-code-agents
Or copy agents manually:
git clone https://github.com/sakuga-cmd/claude-code-agents.git
cp claude-code-agents/agents/*.md ~/.claude/agents/
| Agent | Role | Model | Country | Description |
|---|---|---|---|---|
| Dima | Architect | opus | 🇷🇺 Russia | System design, implementation plans, task decomposition |
| Pierre | Code Reviewer | opus | 🇫🇷 France | Code review, PR review, refactoring suggestions |
| Avi | Security Auditor | opus | 🇮🇱 Israel | OWASP, XSS/CSRF/injection, access policy audit |
| Max | Impact Analyst | opus | 🇨🇭 Switzerland | Impact analysis, hidden dependencies, risk assessment |
| Tomek | Debugger | opus | 🇵🇱 Poland | Root cause analysis, error tracing, scientific debugging |
| Astrid | A11y Auditor | opus | 🇸🇪 Sweden | WCAG compliance, ARIA, keyboard navigation |
| Mark | Code Explorer | sonnet | 🇩🇪 Germany | Codebase research, dependency mapping, context scouting |
| Bas | Web Researcher | sonnet | 🇳🇱 Netherlands | Web search, documentation lookup, library comparison |
| Ron | SEO Specialist | sonnet | 🇺🇸 USA | Meta tags, Open Graph, structured data, Core Web Vitals |
| Jun | Performance Engineer | sonnet | 🇰🇷 Korea | Bundle analysis, Core Web Vitals, rendering optimization |
| Agent | Role | Model | Country | Description |
|---|---|---|---|---|
| Yuki | DBA | opus | 🇯🇵 Japan | Schema design, migrations, access policies, query optimization |
| Timur | Frontend Developer | sonnet | 🇹🇷 Turkey | React components, hooks, state management, routing |
| Oleg | Backend Developer | sonnet | 🇺🇦 Ukraine | API routes, serverless functions, server-side logic |
| Chiara | UI/UX Designer | sonnet | 🇮🇹 Italy | UI components, design system, animations, responsive layouts |
| Aino | Tester | sonnet | 🇫🇮 Finland | TDD, unit/integration tests, edge cases |
| Pablo | Refactoring Specialist | sonnet | 🇪🇸 Spain | Code simplification, dead code removal, deduplication |
| Kate | DevOps Engineer | sonnet | 🇬🇧 UK | Deployment, CI/CD, secrets management, monitoring |
| Lu | PWA/Mobile Specialist | sonnet | 🇧🇷 Brazil | Service worker, push notifications, offline mode |
Personalities with national flavor — Not generic templates. Pierre reviews code like a French sommelier evaluating wine — frank but refined. Avi delivers security findings with Israeli directness — "You have a hole. Fix it now." Pablo simplifies code with Spanish elegancia — "Three lines is a poem. Thirty lines is bureaucracy."
Team awareness — Every agent knows the full 18-person roster. When they need help, they say who to call and why. Handoff notes include a personality-flavored message for the boss.
Read-only safety — Review/audit agents have disallowedTools: Write, Edit in frontmatter. They physically cannot modify code during review.
Structured output — Every agent has a defined output format for consistent, parseable results.
Anti-patterns — Every agent includes BAD/GOOD examples showing what NOT to do, not just what to do.
Handoff notes — Each agent ends with a structured handoff: status, result, who's next, warnings, and a note to the boss in character.
Agents are generic by design — no framework-specific references. To adapt for your stack:
This is an actively developed project. Coming soon:
npx claudepluginhub sakuga-cmd/claude-code-agents --plugin claude-code-agentsAn engineering team in a box for Claude Code. 12 specialized subagents (planner, fullstack-engineer, refactor-specialist, migration-engineer, frontend-designer, critic, vuln-verifier, debugger, db-expert, onboarder, tool-expert, web-researcher) plus 15 automation hooks (pre-commit secret scan, MCP health tracking, cost tracking, test runner, branch protection, large file warner, session summary, batch format, design quality, config protection, and more) wired by the P7/P9/P10 methodology with three red lines: closure discipline, fact-driven, exhaustiveness.
Production-ready Claude Code configuration with role-based workflows (PM→Lead→Designer→Dev→QA), safety hooks, 44 commands, 19 skills, 8 agents, 43 rules, 30 hook scripts across 19 events, auto-learning pipeline, hook profiles, and multi-language coding standards
Expert agents for specific programming languages (Python, Go, Rust, etc.)
Full-stack agents — frontend, backend, API, DevOps architects
5つの専門エージェント(backend-reviewer, frontend-reviewer, test-reviewer, security-reviewer, ux-reviewer)がチームでコードレビューするスキル。
Your AI development team. 36 agents and 32 skills for building, reviewing, debugging, and shipping production code. Supports Vue, React, Next.js, SvelteKit, Angular, Astro, and Nuxt.