KiCad MCP Server
A Model Context Protocol (MCP) server that bridges AI assistants like Claude with KiCad, enabling natural language control and automation of PCB design workflows.
Overview
KiCad MCP Server is an open-source implementation of the Model Context Protocol (MCP) that allows Large Language Models (LLMs) such as Anthropic's Claude to directly interact with KiCad, the popular open-source EDA tool for printed circuit board (PCB) design.
Built on the MCP specification, it exposes KiCad's capabilities—including project management, schematic analysis, layout operations, design rule checking (DRC), and export functions—through a standardized, secure interface. This enables engineers and designers to control complex PCB workflows using natural language instead of manual GUI interactions or scripting.
Features
- Natural Language PCB Control: Instruct AI to open projects, modify schematics, route traces, or generate outputs via conversational commands.
- Real-time Project Access: Read and manipulate current project state, including nets, components, and board layouts.
- Design Rule Checking (DRC): Run automated DRC and receive detailed violation reports.
- Schematic & Layout Analysis: Extract netlists, analyze connectivity, visualize boards, and troubleshoot designs.
- Automation Workflows: Generate BOMs, export Gerbers, perform simulations, or iterate on designs iteratively with AI guidance.
- Cross-Platform Support: Works on Windows, macOS, and Linux with KiCad 9.x and compatible MCP clients (Claude Desktop, VS Code with Cline, etc.).
- Secure Tool Schemas: Adheres to MCP standards for safe, controlled access to KiCad's Python scripting API.
Use Cases
- AI-Assisted PCB Design: Let Claude suggest component placements, optimize routing, or fix DRC errors in real time.
- Rapid Prototyping: Automate repetitive tasks like footprint assignment, netlist generation, or manufacturing file exports.
- Design Review & Validation: Have AI analyze schematics for errors, check manufacturability, or validate against specifications.
- Educational & Learning: New users can query the AI for step-by-step guidance while working inside KiCad.
- Multi-Tool Orchestration: Combine with other MCP servers for end-to-end hardware development pipelines.
Getting Started
- Install KiCad (version 9.x recommended).
- Set up the MCP server via pip (for Python variants) or npm/clone the repository.
- Configure your MCP client (e.g., add to Claude Desktop
server.jsonor equivalent). - Start chatting with your AI about your KiCad project!
Multiple community implementations exist (e.g., lamaalrajih/kicad-mcp, bunnyf/pcb-mcp), offering variations in language (Python/TypeScript) and feature focus.
Compatibility
- MCP Clients: Claude Desktop, Cline (VS Code), and any compliant MCP host.
- KiCad Version: Primarily tested with KiCad 9.x and its Python scripting interface.
- License: Typically MIT or similar open-source terms (check specific repo).
This tool represents a significant step toward AI-native electronics design, reducing manual effort and accelerating hardware iteration cycles.