Openroad Mcp

Name: Openroad Mcp
Author: luarss

@luarss

代码仓库

关于 Openroad Mcp

暂无概览

基本信息

分类

其他

传输方式

stdio

发布者

luarss

提交者

Song Luar

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "openroad-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "git+https://github.com/luarss/openroad-mcp",
        "openroad-mcp"
      ]
    }
  }
}

工具

Execute commands in an interactive OpenROAD session

Create a new OpenROAD session

List all active sessions

Terminate a session

Get detailed session information

View command history

Get performance metrics

List ORFS report directory images

Read a ORFS report image

概览

What is Openroad Mcp?

Openroad Mcp is a Model Context Protocol (MCP) server that provides tools for interacting with OpenROAD and ORFS (OpenROAD Flow Scripts). It enables AI assistants to execute OpenROAD commands in persistent sessions, manage multiple sessions, access command history and performance metrics, and visualize ORFS reports.

How to use Openroad Mcp?

Install via uvx using the standard MCP client JSON configuration: set the command to uvx with arguments --from git+https://github.com/luarss/openroad-mcp openroad-mcp. For Claude Code, add the server via the CLI: claude mcp add --transport stdio openroad-mcp -- uvx --from git+https://github.com/luarss/openroad-mcp openroad-mcp. Once configured, restart the client and tools such as interactive_openroad and create_interactive_session become available.

Key features of Openroad Mcp

Interactive OpenROAD sessions with PTY support
Session creation, listing, inspection, and termination
Command history access for any session
Performance metrics across all sessions
Report image listing and reading from ORFS runs

Use cases of Openroad Mcp

Running and debugging OpenROAD commands interactively via an AI assistant
Managing multiple concurrent OpenROAD sessions in EDA workflows
Monitoring performance metrics across OpenROAD design runs
Inspecting and reading ORFS report images without leaving the MCP client
Automating RTL-to-GDS flow tasks through MCP tools

FAQ from Openroad Mcp

What are the system requirements?

OpenROAD must be installed and available in your PATH. Python 3.13+ and the uv package manager are also required. ORFS (OpenROAD Flow Scripts) is optional but recommended for complete flows.

Which MCP clients are supported?

Claude Code and Gemini CLI are fully supported via STDIO transport. Other MCP clients using STDIO should work with standard configuration.

How do I troubleshoot if the server fails to start?

Ensure uv is installed and in PATH, verify the path to openroad-mcp is correct, run make sync to install all dependencies, and check your MCP client logs for specific error messages.