Anthropic API Server (TypeScript) MCP Server

@ptitiwat2543

代码仓库

基本信息

分类

开发工具

许可证

MIT license

传输方式

stdio

作者

ptitiwat2543

关于 Anthropic API Server (TypeScript) MCP Server

暂无概览

#anthropic-api #typescript #api-server

Anthropic API Server (TypeScript) MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "mcp-p-anthropic-api-server": {
      "command": "npx",
      "args": [
        "mcp-p-anthropic-api-server"
      ]
    }
  }
}

Anthropic API Server (TypeScript) MCP Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。

点击上方按钮,在临时沙箱中运行此服务器并实时获取工具列表。

概览

What is Anthropic API Server (TypeScript)?

Anthropic API Server (TypeScript) is a TypeScript-based MCP server that provides a secure interface for connecting to Anthropic’s Claude API. It gives developers a ready‑to‑use backend with high test coverage, streaming support, function calling, and a clean REST API.

How to use Anthropic API Server (TypeScript)?

Clone the repository, install dependencies (npm install), copy .env.example to .env, and set your ANTHROPIC_API_KEY. Run in development with npm run dev, build and start for production with npm run build then npm start, or invoke directly with npx mcp-p-anthropic-api-server. Use the exposed endpoints (GET /api/health, POST /api/messages, etc.) to interact with Claude.

Key features of Anthropic API Server (TypeScript)

Connects to Anthropic API using the official SDK
Supports streaming responses for real‑time output
Enables function calling / tool usage by Claude
Validates input with Zod schemas
Provides a built‑in logging system
Can be run instantly via npx

Use cases of Anthropic API Server (TypeScript)

Build AI‑powered chatbots and assistants using Claude
Create custom tools that Claude can invoke through the /api/tools/execute endpoint
Integrate Claude’s streaming completions into web or desktop applications
Use with Claude Desktop by pointing to the server in claude_desktop_config.json

常见问题

Why can’t I connect to the Anthropic API?

Check that your `ANTHROPIC_API_KEY` in the `.env` file is correct, verify your internet connection, and try the direct API example at `examples/direct-anthropic-api.js`.

I get “TypeError: app.address is not a function” when running tests.

This occurs with `supertest` when the app object isn’t a valid Express application. Use direct mock tests instead (see `api-coverage-fixed.test.ts`) or intercept route handlers directly.

How do I add a new tool?

Edit `src/services/tools.ts` and call `registerTool()` with the tool’s schema and an async handler. An example for a `getWeather` tool is provided in the README.

What runtime does the server require?

It requires Node.js and npm. TypeScript compilation is handled by the included `tsconfig.json`.

How do I run the test suite?

Use `npm test` for all tests, `npm run test:validation` for schema validation tests, `npm run test:integration` for integration tests, or `npm run test:coverage` for a coverage report.

开发工具分类下的更多 MCP 服务器

Modelglass

Modelglass turns "which model should I use?" into a lookup instead of a guess. Seven tools give your agent live pricing across image, language, video, and audio models. Sourced and versioned, not scraped once and forgott

Ramen

bkraad47

Multi-zone MCP server for GKE/EKS/AKS: Python tools from git, NeMo guardrails, SSO, autoscaling.

Yocoolab

MCP server for AI Agents — exposes Yocoolab feedback threads and design selections as tools

Agent Conductor

icohangar-ops

Agent conductor — AGENTS.md contract parser, SKILL.md registry, CHP decision engine over MCP

Perfex CRM

themesic

Turn Perfex CRM into an AI-ready workspace. This MCP server exposes your full REST API to Claude, ChatGPT and any AI agent, so leads, invoices and tasks are one prompt away.