One ruleset for Cursor, Claude Code, Codex, Antigravity, Q, and Kiro.
6 config files. Always out of sync.
Single source. Auto-synced to all tools.
One source of truth automatically applied to all 6 AI coding tools.
Specialist agents for architecture, security, testing, performance, and accessibility.
PLAN → ACT → EVAL cycle ensures consistent quality across all development.
TDD, SOLID principles, and 90%+ test coverage enforced as standard practice.
Standard Model Context Protocol for seamless AI tool integration.
One command to install. Works out of the box with all supported tools.
+ any MCP-compatible tool
Watch your AI agents work in real-time with a built-in terminal UI.

Visual pipeline showing active agents, stages, and real-time progress
Live view of the active agent with sparkline activity and progress bar
Task completion tracking synced from PLAN/ACT/EVAL context
Real-time tool invocation bar chart updated as agents work
npx codingbuddy tuiMultiple Claude Code sessions share a single TUI window via IPC
Focused expertise for every aspect of development
29 agents
Specialized for work planning and design
Low-level implementation planning with TDD and bite-sized tasks
Architecture expert for Planning, Implementation, and Evaluation
High-level system design and architecture planning specialist
Cross-cutting migration coordinator for legacy system modernization
Specialized for actual implementation execution
Modern React/Next.js specialist with Server Components, TDD, and accessibility
Language-agnostic backend specialist with Clean Architecture and TDD
Get started in 3 simple steps
Add codingbuddy-rules to your project
npx codingbuddy initAdd the MCP server to your AI tool
{
"mcpServers": {
"codingbuddy": {
"command": "npx",
"args": ["-y", "codingbuddy"]
}
}
}Use PLAN, ACT, EVAL modes with specialist agents
PLAN design auth feature
# 🏛️ architecture → System design
# 🧪 test-strategy → TDD approach
# 🔒 security → OWASP checks