Railway MCP Server

Railway MCP Server

Enables management of Railway.app infrastructure through natural language, allowing users to deploy services, manage environment variables, monitor deployments, and configure projects directly from MCP clients like Claude.

Category
访问服务器

README

Railway MCP Server

<p align="center"> <picture> <source media="(prefers-color-scheme: dark)" srcset="https://avatars.githubusercontent.com/u/66716858?s=200&v=4" /> <source media="(prefers-color-scheme: light)" srcset="https://avatars.githubusercontent.com/u/66716858?s=200&v=4" /> <img alt="Railway" src="https://avatars.githubusercontent.com/u/66716858?s=200&v=4" height="40" /> </picture>    <picture> <source media="(prefers-color-scheme: dark)" srcset="https://github.com/user-attachments/assets/38db1bcd-50df-4a49-a106-1b5afd924cb2" /> <source media="(prefers-color-scheme: light)" srcset="https://github.com/user-attachments/assets/82603097-07c9-42bb-9cbc-fb8f03560926" /> <img alt="MCP" src="https://github.com/user-attachments/assets/82603097-07c9-42bb-9cbc-fb8f03560926" height="40" /> </picture> </p>

<p align="center"> <strong> Let Claude and other MCP clients manage your Railway.app infrastructure. Deploy services, manage variables, and monitor deployments - all through natural language. </strong> </p> <p align="center"> <table align="center"> <th align="center"> Please Note: This is under development and not all features are available yet. 🚧 </th> </table> </p>

<a href="https://glama.ai/mcp/servers/lwn74iwigz"> <img width="380" height="200" src="https://glama.ai/mcp/servers/lwn74iwigz/badge" alt="railway-mcp MCP server" /> </a>

A Model Context Protocol (MCP) server for integrating with the Railway.app platform.

smithery badge

MseeP.ai Security Assessment Badge

Verified on MseeP

Table of Contents

<p align="center"> <a href="#features">Features</a> • <a href="#installation">Installation</a> • <a href="#available-tools">Available Tools</a> • <a href="#example-workflows">Example Workflows</a> • <a href="#security-considerations">Security</a> • <a href="#troubleshooting">Troubleshooting</a> • <a href="#contributing">Contributing</a> </p>

Features

Status Meaning
Complete
🚧🔨⏳ Being Built or Needs Testing
Not Built at the moment
  • ✅ Authentication with Railway API tokens
  • ✅ Project management (list, info, delete)
  • ✅ Deployment management (list, restart)
  • ✅ Service management (create from GitHub repo or Docker image, list)
  • ✅ Variable management (list, create/update, delete)
  • ✅ Service Network management
  • ✅ Volume management
  • ❌ Full support for all templates
    • 🚧🔨⏳ Database template support
    • Automatic database and networking workflows
  • 🚧🔨⏳ Most commonly used workflows
  • ❌ Automatic GitHub repository linking for services

Installation

Prerequisites

  • Node.js 18+ (for built-in fetch API support)
  • An active Railway account
  • A Railway API token (create one at https://railway.app/account/tokens)

Quick Start

This MCP server is designed to work with MCP Clients like:

  • Claude for Desktop | ✅ Battle-Tested
  • Cursor | ✅ Needs Testing
  • Cline | 🚧🔨⏳ Needs Testing
  • Windsurf | 🚧🔨⏳ Needs Testing
  • Other MCP Clients | 🚧🔨⏳ Needs Testing

Installing via Smithery

To install railway-mcp automatically, we recommend using Smithery

Claude Desktop

npx -y @smithery/cli install @jason-tan-swe/railway-mcp --client claude

Cursor

npx -y @smithery/cli@latest run @jason-tan-swe/railway-mcp --config "{\"railwayApiToken\":\"token\"}"

<details> <summary> <h3>Manual Installation For Cursor</h3></summary>

  1. Head to your cursor settings and find the MCP section

  2. Click 'Add new MCP server'

  3. Name it however, you like, we recommend railway-mcp for better clarity

  4. Paste this command into the 'Command' section, where <RAILWAY_API_TOKEN> is your accounts Railway token:

npx -y @jasontanswe/railway-mcp <RAILWAY_API_TOKEN>

</details>

<details>

<summary><h3>Manual Installation For Claude</h3></summary>

  1. Create or edit your Claude for Desktop config file:

    • macOS: ~/Library/Application\ Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json
  2. Add the railway-mcp server to your configuration with your API token:

   "railway": {
      "command": "npx",
      "args": ["-y", "@jasontanswe/railway-mcp"],
      "env": {
         "RAILWAY_API_TOKEN": "your-railway-api-token-here"
      }
   }

When you have multiple MCP servers, your config file might look like this:

{
  "mcpServers": {
    // ... All of your existing MCP servers ...

    // Add the railway-mcp server to your configuration with your API token
    "railway": {
      "command": "npx",
      "args": ["-y", "@jasontanswe/railway-mcp"],
      "env": {
        "RAILWAY_API_TOKEN": "your-railway-api-token-here"
      }
    }
  }
}
  1. Restart Claude for Desktop

  2. You can now start using Railway tools directly in Claude. For example:

Please list all my Railway projects
  1. Alternatively, if you don't want to add your token to the configuration file, you can configure it within Claude using:
Please configure the Railway API with my token: {YOUR_API_TOKEN_HERE}

</details>

Recommendations and Other Information

This server best combines with MCP-clients that have access to terminal or with Git (Cursor, Windsurf). Using this MCP with others is recommended as railway-mcp orchestrates containers and streamlines your deployment process seamlessly.

Recommended MCP servers to combine with

For Claude

  • Out of the box, Claude does not have terminal access, so it cannot trigger deployments as it will not be able to get the latest commit.
  • Spinning up different services and monitoring them are the best use case with Claude.

For Cursor

  • Use with GitHub MCP or have the repository already setup on GitHub and cloned locally on your machine to leverage full integration with railway-mcp.
  • When Cursor makes a change, it may forget to push it's changes to GitHub causing it to try and deploy a commit that Railway cannot pull.
    • SOLUTION: Always ask or include somewhere in your prompt: Have you pushed our changes to GitHub yet?

Security Considerations

  • Railway API tokens provide full access to your account. Keep them secure.
  • When using the environment variable method, your token is stored in the Claude Desktop configuration file.
  • Sensitive variable values are automatically masked when displayed.
  • All API calls use HTTPS for secure communication.
  • The server's memory-only token storage means your token is never written to disk outside of the configuration file.

Troubleshooting

If you encounter issues:

  1. Token Authentication Issues

    • Ensure your API token is valid and has the necessary permissions
    • If using the environment variable method, check that the token is correctly formatted in the config file
    • Try using the configure tool directly in Claude if the environment token isn't working
  2. Server Connection Issues

    • Check that you've installed the latest version of the server
    • Verify that Node.js version 18 or higher is installed
    • Restart Claude for Desktop after making changes to the configuration
  3. API Errors

    • Verify that you're using correct project, environment, and service IDs
    • Check Railway's status page for any service disruptions
    • Railway API has rate limits - avoid making too many requests in a short period

Contributing

We welcome contributions from the community! Please see our Contributing Guidelines for details on how to get started, development guidelines, and debugging information.

</details>

Available Tools

<details> <summary>View All</summary>

Authentication

  • configure - Set your Railway API token (only needed if not provided in environment variables)

Projects

  • project-list - List all projects in your account
  • project-info - Get detailed information about a specific project
  • project-create - Create a new project with optional team ID
  • project-delete - Delete a project
  • project-environments - List all environments in a project

Services

  • service-list - List all services in a specific project
  • service-info - Get detailed information about a specific service
  • service-create-from-repo - Create a new service from a GitHub repository
  • service-create-from-image - Create a new service from a Docker image
  • service-delete - Delete a service from a project
  • service-restart - Restart a service in a specific environment
  • service-update - Update service configuration (build command, start command, etc.) | 🚧 Needs Testing

Deployments

  • deployment-list - List recent deployments for a service
  • deployment-trigger - Trigger a new deployment for a service
  • deployment-logs - Get logs for a specific deployment
  • deployment-health-check - Check the health/status of a deployment

Variables

  • variable-list - List variables for a service or environment
  • variable-set - Create or update a variable
  • variable-delete - Delete a variable
  • variable-bulk-set - Bulk update variables for a service | 🚧 Needs Testing
  • variable-copy - Copy variables between environments | 🚧 Needs Testing

Databases

  • database-list-types - List all available database types that can be deployed
  • database-deploy - Deploy a new database service </details>

<details> <summary>Example Workflows</summary>

Setting up a new service

  1. List projects to get the project ID
  2. Create a new service from a template
  3. Add environment variables
  4. View the service deployment

Managing environment variables

  1. List projects to find your project ID
  2. List variables to see what's currently set
  3. Create or update variables as needed
  4. Delete any obsolete variables

</details>

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选