Everything OpenCode

A comprehensive collection of agents, commands, skills, and configurations for OpenCode - the open-source AI coding agent.

Overview

This repository provides a ready-to-use toolkit for enhancing your OpenCode workflow with:

• 9 Specialized Agents - For planning, architecture, TDD, code review, security, and more
• 15 Custom Commands - Slash commands for common development workflows
• 11 Skills - Reusable knowledge patterns for backend, frontend, security, and testing
• MCP Server Configs - Pre-configured Model Context Protocol servers

Quick Start

1. Clone or Copy

# Clone to use as a template
git clone https://github.com/your-username/everything-opencode.git

# Or copy files to your project
cp -r .opencode/ /path/to/your/project/
cp AGENTS.md opencode.jsonc /path/to/your/project/

2. Configure

Edit opencode.jsonc to set your environment variables:

{
  "mcp": {
    "github": {
      "environment": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "${env:GITHUB_PAT}"
      }
    }
  }
}

3. Use

# Start OpenCode in your project
opencode

# Use slash commands
/plan Add user authentication
/tdd Create login function
/code-review
/verify

Project Structure

everything-opencode/
├── .opencode/
│   ├── agents/           # Agent definitions
│   │   ├── architect.md
│   │   ├── build-error-resolver.md
│   │   ├── code-reviewer.md
│   │   ├── doc-updater.md
│   │   ├── e2e-runner.md
│   │   ├── planner.md
│   │   ├── refactor-cleaner.md
│   │   ├── security-reviewer.md
│   │   └── tdd-guide.md
│   │
│   ├── commands/         # Slash commands
│   │   ├── build-fix.md
│   │   ├── checkpoint.md
│   │   ├── code-review.md
│   │   ├── e2e.md
│   │   ├── eval.md
│   │   ├── learn.md
│   │   ├── orchestrate.md
│   │   ├── plan.md
│   │   ├── refactor-clean.md
│   │   ├── setup-pm.md
│   │   ├── tdd.md
│   │   ├── test-coverage.md
│   │   ├── update-codemaps.md
│   │   ├── update-docs.md
│   │   └── verify.md
│   │
│   └── skills/           # Reusable skills
│       ├── backend-patterns/
│       ├── clickhouse-io/
│       ├── coding-standards/
│       ├── continuous-learning/
│       ├── eval-harness/
│       ├── frontend-patterns/
│       ├── project-guidelines-example/
│       ├── security-review/
│       ├── strategic-compact/
│       ├── tdd-workflow/
│       └── verification-loop/
│
├── agents/               # Original Claude Code agents (reference)
├── commands/             # Original Claude Code commands (reference)
├── rules/                # Original rules files (reference)
├── skills/               # Original skills (reference)
│
├── AGENTS.md             # Project instructions for OpenCode
├── opencode.jsonc        # Main configuration file
└── plugins/README.md     # MCP servers documentation

Available Agents

Agent	Purpose	When to Use
planner	Implementation planning	Complex features, refactoring
architect	System design	Architectural decisions
tdd-guide	Test-driven development	New features, bug fixes
code-reviewer	Code review	After writing code
security-reviewer	Security analysis	Before commits
build-error-resolver	Fix build errors	When build fails
e2e-runner	E2E testing	Critical user flows
refactor-cleaner	Dead code cleanup	Code maintenance
doc-updater	Documentation	Updating docs

Available Commands

Command	Description
/plan	Create implementation plan before coding
/tdd	Enforce test-driven development workflow
/code-review	Security and quality review of changes
/build-fix	Fix TypeScript and build errors
/verify	Run comprehensive verification checks
/e2e	Generate and run Playwright E2E tests
/checkpoint	Create/verify workflow checkpoints
/eval	Eval-driven development workflow
/learn	Extract reusable patterns as skills
/orchestrate	Sequential multi-agent workflows
/refactor-clean	Identify and remove dead code
/test-coverage	Analyze and improve test coverage
/update-codemaps	Update architecture documentation
/update-docs	Sync documentation from source
/setup-pm	Configure package manager

Available Skills

Skill	Description
backend-patterns	API design, database optimization, caching
frontend-patterns	React, Next.js, state management, performance
tdd-workflow	Test-driven development methodology
security-review	Security checklist and patterns
coding-standards	TypeScript, React, Node.js best practices
clickhouse-io	ClickHouse analytics patterns
eval-harness	Eval-driven development framework
verification-loop	Comprehensive verification system
strategic-compact	Context management optimization
continuous-learning	Extract patterns from sessions
project-guidelines-example	Template for project-specific skills

MCP Servers

Pre-configured MCP servers in opencode.jsonc:

Server	Type	Purpose
GitHub	local	PRs, issues, repos
Memory	local	Persistent memory
Context7	local	Live documentation
Sequential Thinking	local	Chain-of-thought reasoning
Firecrawl	local	Web scraping
Supabase	local	Database operations
Vercel	remote	Deployments
ClickHouse	remote	Analytics queries
Cloudflare	remote	Workers and docs

Configuration

AGENTS.md

Project-wide instructions that OpenCode reads automatically:

• Coding style guidelines
• Security requirements
• Testing requirements
• Git workflow
• Common patterns

opencode.jsonc

Main configuration file with:

{
  "$schema": "https://opencode.ai/config.schema.json",
  
  // MCP server configurations
  "mcp": { ... },
  
  // Agent definitions (alternative to .opencode/agents/)
  "agent": { ... },
  
  // Tool permissions
  "permissions": { ... },
  
  // Project settings
  "project": { ... }
}

Customization

Adding Custom Agents

Create a new file in .opencode/agents/:

---
description: Your agent description
mode: subagent
model: anthropic/claude-sonnet-4-20250514
---

Your agent prompt here...

Adding Custom Commands

Create a new file in .opencode/commands/:

---
description: Command description shown in /help
---

Command instructions here...

Adding Custom Skills

Create a new directory in .opencode/skills/your-skill/:

---
name: your-skill
description: When to use this skill
---

Skill content here...

Best Practices

1. Start with /plan - Always plan before implementing complex features
2. Use /tdd - Write tests before implementation
3. Run /code-review - Review code before committing
4. Check /verify - Ensure all quality gates pass before PRs
5. Invoke agents proactively - Don't wait for the user to ask

Contributing

1. Fork the repository
2. Create a feature branch: git checkout -b feat/new-agent
3. Make your changes
4. Run verification: /verify
5. Commit: git commit -m "feat(agents): add new agent"
6. Push and create PR

Resources

• OpenCode Documentation
• MCP Specification
• Anthropic Claude API

License

MIT License - See LICENSE for details.