pcb-mcp
Enables AI assistants to read and modify KiCAD PCB designs through the KiCAD IPC API, providing tools for board queries, footprint placement, track creation, DRC, and export.
README
pcb-mcp
An MCP server that lets an AI assistant read and modify KiCAD PCB designs through KiCAD's IPC API.
Status
Pre-alpha. There is no working code yet. This is actively being built as a personal learning project and is not usable today. It is not installable.
What it will do
The server exposes a small set of tools (20 or fewer) so an AI assistant can pick the right one reliably:
- Query board state (layers, nets, footprints, tracks, zones)
- Place and move footprints
- Create tracks and vias
- Run Design Rule Check (DRC)
- Export manufacturing files (Gerbers, drill files, BOM)
Design decisions
| Choice | Reason |
|---|---|
| Python 3.11+ | Matches KiCAD plugin ecosystem, broad library support |
Official mcp SDK v2 (MCPServer, stdio transport) |
Canonical implementation, stable as of 2026-07-28 |
kicad-python over the KiCAD IPC API |
MIT-licensed, actively maintained, replaces deprecated SWIG pcbnew bindings |
| KiCAD 10.0.x target | Current stable (10.0.5). IPC API first shipped in KiCAD 9 but matured in 10 |
| PCB editor scope only | The IPC API has no schematic access today |
| MIT license (out-of-process client) | No GPL entanglement because the server communicates over IPC, not linked to KiCAD |
Requirements
- KiCAD 10.0.x installed and running with a board open. The IPC API needs a live GUI instance (headless mode is expected in KiCAD 11).
- KiCAD API enabled in Preferences → API settings.
- Python 3.11+
- Only one IPC client can attach to a KiCAD instance at a time.
Development
This project follows a branch-and-pull-request workflow. The main branch is protected, and every change lands through a PR.
Detailed planning and research documents are kept locally and are not published in this repo. The README is the single source of project context for external readers.
License
MIT. See LICENSE.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。