namegen-mcp
Generates fantasy names for characters, towns, regions, and more, backed by the @ironarachne/made-up-names dataset. It exposes a single tool, generate-names, with parameters for race, type, and number.
README
namegen-mcp
An MCP server that mirrors the behaviour of https://names.ironarachne.com, backed by the @ironarachne/made-up-names dataset. It exposes a single tool, generate-names, which accepts race, type, and number arguments to return randomly generated names. The defaults match the hosted API (race defaults to human, number defaults to 1).
Prerequisites
- Node.js 18 or newer
- npm 9 or newer (ships with recent Node.js distributions)
Installation
npm install
This downloads the MCP SDK and the name generation dataset locally so the server operates without any external network calls.
Running the server
Development (hot reload)
npm run dev
tsx compiles the TypeScript entry point in-memory and wires the server to stdio, which is the transport expected by most MCP-compatible clients.
Production build
npm run build
npm start
This emits the compiled files into dist/ and runs the stdio server from the generated JavaScript.
Using with Claude Desktop (and other MCP frontends)
-
Open Claude Desktop and navigate to Settings → Developer → Model Context Protocol.
-
Add a new MCP server entry with one of the following command setups (update
cwdto wherever you cloned the repo). Claude’s configuration expects servers inside amcpServersobject, so the full JSON looks like this:{ "mcpServers": { "namegen": { "command": "npm", "args": ["start"], "cwd": "/absolute/path/to/namegen-mcp" } } }To run the TypeScript entry without a prior build, swap the command block for the following:
{ "mcpServers": { "namegen": { "command": "npx", "args": ["tsx", "src/server.ts"], "cwd": "/absolute/path/to/namegen-mcp" } } }Important: Set
cwdto the project root (the folder containingpackage.json). Pointing it atdist/or another subdirectory will makenpmfail to locate the scripts file and the server exits with code 1. If you use the Settings UI, Claude writes this JSON for you; if you editclaude-desktop-config.jsonmanually, be sure to keep themcpServerswrapper or the app will report “Invalid config: missing mcpServers object.” -
For the
npm startvariant, runnpm run buildonce sodist/server.jsis available. Thenpx tsxcommand path does not require a pre-build. -
Save the configuration and restart Claude Desktop if it does not immediately recognise the new server.
Any other MCP-compatible frontend that supports stdio transports can use the same command/args/cwd configuration to launch this project.
Tool reference
- Tool ID:
generate-names - Parameters:
race(string, optional, default"human")type(string, required, one ofmale,female,family,given,region,town,culture)number(integer, optional, default1, maximum50)
- Response: The tool returns a JSON payload with the canonical
race, requestedtype, requestedcount, and an array of generatednames. The same payload is returned as structured content and rendered text for convenience.
Two static resources are also registered:
namegen://races– JSON list of valid race identifiersnamegen://name-types– JSON list of valid name types
Example
Requesting three female elf names via an MCP client might return a payload like the following:
{
"race": "elf",
"type": "female",
"count": 3,
"names": [
"Aerithain",
"Nylenna",
"Valesri"
]
}
Because all content is generated locally, the server can run offline after the initial dependency installation.
Troubleshooting
- Server exits immediately with code 1 in Claude: make sure the
cwdin your MCP configuration points to the repository root (wherepackage.jsonlives) and that you rannpm install. If you use thenpm startcommand, runnpm run buildonce sodist/server.jsexists.
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。