Complete setup guide for the Agent Kit — a portable multi-stack AI agent system for OpenCode. Includes 33 specialized agents, 207 skill playbooks, 47 slash commands, 10 MCP servers, and an automatic visual dev loop for iterative UI development.
Multi-stack OpenCode agent toolkit — 33+ specialized AI agents, 200+ skills, 46 commands, 8 MCP servers (Nuxt, React, Node.js, Laravel, CI3, Android, Flutter, DevOps, SEO, SonarQube, and more)
2026-11-16
Sterne
—
Tools
—
Uptime
—
npx opencode-agent-kit init # One command. Full team.
Quick Install
Per-Project Install (Traditional)
npx opencode-agent-kit init
Copies the full agent configuration, skills, commands, and rules into .opencode/ in your project.
Global Install (Recommended — All Projects)
Install once, use in every project. OpenCode auto-detects the global config.
# Step 1: Install globally (one time)
npx opencode-agent-kit init --global
# Step 2: Just use opencode in any project — no per-project setup needed
cd /path/to/your-project
opencode
The global install puts everything in the platform-specific OpenCode config directory:
# Check global install status
opencode-agent-kit global status
# Show global install path
opencode-agent-kit global path
# Update to latest version
opencode-agent-kit global update
# Link a project explicitly (macOS/Linux: auto-detected)
opencode-agent-kit link
# Force per-project copy (detach from global)
npx opencode-agent-kit init --local
Overview
Think of Agent Kit as an AI development team you install into any OpenCode project. An IT Leader orchestrates 33 specialized agents — each one an expert in their stack. You describe what to build; the team handles the rest.
This repository contains a complete OpenCode agent configuration with Leader → Subagent architecture for software development teams.
Agent Config
Active config: opencode.json (source of truth)
Agent prompts: .opencode/prompts/agents/
it-leader.md — IT Leader & Technical Project Manager (primary)
Structured 30-day mentoring for Nuxt stack transition
Prerequisites
OpenCode CLI installed
Node.js + npm/pnpm/yarn/bun
Integration with Built-in Agents
OpenCode has built-in agents available globally. This KIT does not redefine existing agents — it focuses on specialized agents for each development stack.
Built-in Agents (Global)
Built-in Agent
Model
Role
Usage
planner
opus-4.5
Detailed planning, architectural decisions
/plan or @planner
architect
opus-4.5
System design, scalability analysis
@architect
code-reviewer
opus-4.5
Quality, security, maintainability review
/code-review or @code-reviewer
security-reviewer
opus-4.5
Vulnerability detection
/security or @security-reviewer
tdd-guide
opus-4.5
TDD workflow, 80%+ coverage enforcement
/tdd or @tdd-guide
build-error-resolver
opus-4.5
Fix TypeScript/build errors
/build-fix or @build-error-resolver
e2e-runner
opus-4.5
Playwright E2E test generation & execution
/e2e or @e2e-runner
refactor-cleaner
opus-4.5
Dead code removal, consolidation
/refactor-clean or @refactor-cleaner
database-reviewer
opus-4.5
PostgreSQL query optimization, Supabase best practices
@database-reviewer
Custom Agents (KIT-specific)
Specialized agents not available in built-in OpenCode:
You describe what you need — a feature, a bug fix, a full app
The IT Leader analyzes requirements, designs architecture, and breaks the work into tasks
Tasks are delegated to the right subagents — frontend, backend, mobile, designer, reviewer
The IT Leader integrates results and reports back with verification status
For small tasks, mention subagents directly with @mention:
@frontend-nuxt Add a UButton "Save" in ProfileHeader.vue.
@frontend-react Create a server component ProductList with API fetch.
@node-developer Add endpoint POST /api/markets with DTO validation.
@laravel Build a CRUD product page with Blade + Livewire.
@ci3 Build a REST API for products with JWT auth.
@designer Review UX flow for checkout page and generate DESIGN.md.
@reviewer Audit security for the authentication module.
@database Optimize query for market listing with pagination.
@devops Setup CI/CD pipeline for Vercel deployment.
@seo Implement meta tags and structured data for product pages.
@android Build a login screen with Jetpack Compose + ViewModel.
@flutter Build a product list screen with Bloc pattern.
@python Build a FastAPI backend with SQLAlchemy models.
@rust Implement a CLI tool with command parsing and file I/O.
@swift Build an iOS settings screen with SwiftUI.
@dotnet Create an ASP.NET Core Web API with EF Core.
@angular Build a reactive dashboard component with RxJS.
@cpp Implement a performant data processing module.
@agent-engineer Optimize harness hooks and agent routing config.
@python-reviewer Audit Python code for security and patterns.
@rust-reviewer Review Rust code for memory safety.
@go-reviewer Review Go code for concurrency correctness.
@java-reviewer Review Java Spring Boot service layer.
@kotlin-reviewer Review Kotlin coroutine and Compose code.
@php-reviewer Audit PHP code for security vulnerabilities.
@docs Look up the latest API docs for Nuxt 4.
@harness Run cost analysis on the current agent harness.
@loop Start a supervised loop for batch file processing.
Recommended Models
Each agent can use a different model based on task complexity. Subagents inherit the model from the primary agent if not set.
Agent
Role
Recommended Model
Alternative Model
Rationale
IT Leader
Orchestration, architecture, planning
opencode/claude-opus-4.7
opencode/claude-opus-4.5
Deep reasoning, complex analysis, multi-subagent coordination
Frontend
Component implementation, pages, logic
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4.6
Balanced reasoning & efficiency for coding
Backend
API, DTO, controller, database operations
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4.6
Balanced reasoning & efficiency for coding
Designer
Design system, visual exploration, handoff
opencode/claude-sonnet-4
openai/gpt-5
Creativity with sufficient reasoning
Reviewer
Security audit, detailed code review
opencode/claude-opus-4.5
openai/o3
Deep analysis, subtle pattern detection
Database
Schema design, query optimization
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4
Moderate reasoning, precision-focused
DevOps
CI/CD config, scripts, monitoring
opencode/claude-haiku-4.5
openai/gpt-4.1-mini
Straightforward tasks, high efficiency
Android
Kotlin, Compose, Gradle, Play Store
opencode/claude-sonnet-4.5
opencode/claude-haiku-4.5
Balanced reasoning & efficiency
Flutter
Dart, Flutter SDK, cross-platform
opencode/claude-sonnet-4.5
opencode/claude-haiku-4.5
Balanced reasoning & efficiency
Python
Python/FastAPI/Django implementation
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4.6
Balanced reasoning & efficiency for coding
Rust
Rust systems programming, CLI, WebAssembly
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4.6
Balanced reasoning & efficiency for low-level code
Swift
Swift/iOS development, SwiftUI
opencode/claude-sonnet-4.5
opencode/claude-haiku-4.5
Balanced reasoning & efficiency
.NET
C#/.NET backend, ASP.NET Core
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4.6
Balanced reasoning & efficiency for coding
Angular
Angular/TypeScript frontend
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4.6
Balanced reasoning & efficiency for coding
C++
C++ systems programming, performance
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4.6
Balanced reasoning & efficiency for low-level code
Agent Eng
Harness config, evals, routing, cost
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4
Moderate reasoning, configuration-focused
Code Revs
Language-specific code review & audit
opencode/claude-opus-4.5
openai/o3
Deep analysis, subtle pattern detection per language
Docs
Documentation lookup & synthesis
opencode/claude-haiku-4.5
openai/gpt-4.1-mini
Fast lookups, no deep coding
Harness
Harness optimization & cost analysis
opencode/claude-sonnet-4.5
opencode/claude-sonnet-4
Balanced reasoning for config/system work
Loop
Autonomous loop operation & monitoring
opencode/claude-haiku-4.5
openai/gpt-4.1-mini
Efficient, safety-focused monitoring tasks
SEO
Research, meta tags, structured data
openai/gpt-5.1-codex-mini
openai/gpt-5-nano
Research-focused, no deep coding needed
Model Tiers
Tier
Models
Use Case
Tier 1 (Premium)
claude-opus-4.7, claude-opus-4.5
Orchestration, deep analysis, security audit
Tier 2 (Balanced)
claude-sonnet-4.5/4.6, claude-sonnet-4
Complex implementation, design, database
Tier 3 (Efficient)
claude-haiku-4.5, gpt-4.1-mini
Direct tasks, scripts, config files
Tier 4 (Fast)
gpt-5-nano, gpt-5.1-codex-mini
Research, content, cost optimization
Changing Models
All agents — set model at the primary agent (IT Leader) level
Per agent — set model in each agent's config
Quick override — use --model flag when running opencode
See .opencode/config.example.json for a complete per-model configuration example.
Skills
Skills are stored in .opencode/skills/ (local in the repo) — no need to search for them individually.
The following skills are not required for the primary Nuxt + Node.js stack but are now used by new specialized agents (Python, Rust, Go, Java, C++, etc.):
springboot-*, java-*, jpa-patterns
django-*
golang-*
python-*
clickhouse-io
These can be kept if your team uses multi-stack, but are optional.
Agent Kit includes an automatic visual development loop that lets the agent iterate on UI until it's perfect. No manual screenshot requests or "what does it look like?" conversations.
agentmemory is the memory layer for your AI agents. It gives every agent persistent cross-session recall — they remember decisions, insights, and patterns from yesterday's session today.
Why It Matters
Without memory, every session starts from zero. With agentmemory:
53 MCP tools — memory_save, memory_recall, memory_smart_search, memory_sessions, memory_file_history, memory_lesson_save, memory_patterns, and more
Cross-session persistence — Start a new session tomorrow; agents recall yesterday's context, decisions, and code insights
This installs everything to OpenCode's global config directory (~/.config/opencode/ on macOS/Linux, %USERPROFILE%\.config\opencode\ on Windows). OpenCode auto-detects it in every project.
See the Quick Install section for complete details and management commands.
Install rules (follow README in rules/ — copy per directory, don't flatten).
Hooks: only activate if your local environment supports the dependencies. hooks/hooks.json calls scripts via CLAUDE_PLUGIN_ROOT.
Recommended Practices
Use commands/ for routine work
Use rules/ as mandatory standards during implementation and review
Use contexts/ for more precise prompting
Treat hooks/ as team-shared configuration (changes need sync)
Usage
Main Workflow (via IT Leader)
The IT Leader is the primary agent — auto-activated when a session starts. Give it your requirements:
Build a marketplace feature with listing, detail, and create pages.
Backend API for CRUD markets with pagination and filters.
Setup CI/CD pipeline and SEO optimization.
The IT Leader will:
Analyze requirements and define scope
Design architecture (data flow, API contract, component structure)
Break into tasks and delegate to the right subagents
Integrate results and report status
Direct Subagent Calls (small tasks)
For small tasks, mention subagents directly:
@frontend-nuxt Add a UButton "Save" in app/components/profile/ProfileHeader.vue.
Task tiny, minimal diff, don't touch other files.
@frontend-nuxt Implement status filters on the markets page.
Use the existing useApi pattern and report verification status.
@node-developer Add endpoint POST /api/markets with DTO validation.
@designer Create a design system for the product page with Nuxt UI.
@reviewer Audit security for the authentication module.
@database Optimize query for market listing with pagination.
@devops Setup CI/CD pipeline for Vercel deployment.
@seo Implement meta tags and structured data for product pages.
Noch keine Tool-Liste. Remote-Server erhalten eine nach dem ersten Health-Check.
FAQ
Fragen
Wie installiere ich Opencode Agent Kit?
Kopieren Sie auf dieser Seite die stdio-Konfiguration für Cursor, Claude Desktop, Claude Code oder VS Code. Sie nutzt die Installation des Herausgebers, keinen PlugMCP-Proxy.
Benötigt Opencode Agent Kit einen API-Schlüssel?
In diesem Eintrag ist keine erforderliche Umgebungsvariable aufgeführt.