codingbuddy
Codingbuddy orchestrates 29 specialized AI agents to deliver code quality comparable to a team of human experts through a PLAN → ACT → EVAL workflow.
claude mcp add --transport stdio jeremydev87-codingbuddy npx codingbuddy mcp
How to use
Codingbuddy is an MCP server that orchestrates a team of 35 specialized AI agents to collaboratively plan, implement, and verify code. The system uses a three-stage workflow: PLAN to design architecture and test strategy, ACT to implement code following quality standards and TDD, and EVAL to perform multi-dimensional reviews across security, accessibility, performance, and overall quality. You can trigger the workflow via the MCP server to begin a structured development cycle, and you can observe progress through the built-in Terminal Dashboard (TUI) which surfaces real-time agent activity and task state. The tools you’ll interact with include plan-generation, architectural design, backend/frontend development modules, security and performance specialists, and automated code reviews, all coordinated to ship production-ready code when all risk gates are cleared.
How to install
Prerequisites:
- Node.js 18.x or higher
- npm 9.x+ (or yarn 4.x+)
- Access to a shell/terminal with internet
Installation steps:
- Install Codingbuddy globally (if you haven’t already):
npm install -g codingbuddy
- Initialize MCP server configuration (example shown below):
npx codingbuddy init
- Ensure MCP server configuration includes the codingbuddy server, e.g.:
{
"mcpServers": {
"codingbuddy": {
"command": "npx",
"args": ["codingbuddy", "mcp"]
}
}
}
- Start the MCP server (with optional TUI):
npx codingbuddy mcp --tui
- Verify the server is running by checking the console output for the PLAN → ACT → EVAL cycle notifications and agent activity dashboards.
Additional notes
Tips and common considerations:
- The MCP server expects to orchestrate multiple specialized agents; if a particular agent pool is unavailable, the workflow may stall at the PLAN or EVAL stage.
- You can enable the TUI to monitor real-time progress; use --tui with the mcp command.
- Environment variables may be required for integrations or API keys in your project; define VAR_NAME-style placeholders in the env section of your mcp configuration if needed.
- For CLI troubleshooting, consult the Quick Reference and Architecture docs included with Codingbuddy for specific commands and modes.
- If you need to run without npm/npx, you can adapt the workflow to other runners by adjusting the mcpServers entry accordingly (e.g., docker or node executables) but ensure the command and args match your environment.
Related MCP Servers
claude-code-open
Open source AI coding platform with Web IDE, multi-agent system, 37+ tools, MCP protocol. MIT licensed.
sub-agents
Define task-specific AI sub-agents in Markdown for any MCP-compatible tool.
codebase-context
Local-first Second brain for AI agents working on your codebase - detects your team coding conventions and patterns, brings in persistent memory, code-generation checks, and hybrid search with evidence scoring. Exposed through CLI and MCP server.
vibe-check
Stop AI coding disasters before they cost you weeks. Real-time anti-pattern detection for vibe coders who love AI tools but need a safety net to avoid expensive overengineering traps.
lad_mcp_server
Lad MCP Server: Autonomous code & system design review for AI coding agents (Claude Code, Cursor, Codex, etc.). Features multi-model consensus via OpenRouter and context-aware reviews via Serena.
floop
Spreading activation memory for AI coding agents - corrections in, context-aware behaviors out.