MCP Manager

MCP Manager

A gateway that manages multiple MCP servers behind one endpoint, providing namespaced tool routing and supervision. Currently bundles an Arxiv MCP server for paper search and retrieval.

Category
访问服务器

README

MCP Manager

A development scaffold for managing multiple MCP servers behind one standard MCP gateway. The first catalog entry is the open-source Arxiv MCP server.

This repository intentionally separates the components that MCP-Atlas and Toolathlon combine in different ways:

Agent harness
    |
    | Streamable HTTP MCP
    v
Gateway (tool discovery, namespacing, routing)
    |
    v
Runtime supervisor (startup, timeout, retry, cleanup)
    |
    | stdio MCP
    v
Arxiv MCP server
    |
    v
arXiv and related upstream APIs

This is a demo for further development. Dependencies and the Arxiv server are declared but are not vendored into this repository.

Components

Component Location Responsibility
Server catalog catalog/servers/ Declarative, one-file-per-server definitions.
Profiles profiles/ Select servers and allow/deny tools for a workload.
Configuration src/mcp_manager/config/ Typed manifest models and YAML loading.
Secrets src/mcp_manager/secrets/ Resolve explicit secret references without writing resolved config.
Runtime src/mcp_manager/runtime/ Launch and supervise backend MCP connections.
Gateway src/mcp_manager/gateway/ Expose namespaced tools and route calls to backends.
Agent interface src/mcp_manager/harness/ Framework-neutral interface and remote MCP client adapter.

See docs/ARCHITECTURE.md for the design boundaries.

Arxiv backend

The catalog pins blazickjp/arxiv-mcp-server at 0.6.2 and launches it in an isolated uvx environment:

uvx --from arxiv-mcp-server==0.6.2 arxiv-mcp-server

The upstream server uses stdio by default and does not require an API key. Gateway-visible names are stable and namespaced, for example:

arxiv__search_papers
arxiv__get_abstract
arxiv__download_paper

The double-underscore name is stored in a routing table; the gateway never parses it to determine the backend.

Intended development commands

After installing Python 3.11+ and uv:

uv sync --extra dev
uv run mcp-manager list-servers
uv run mcp-manager doctor
uv run mcp-manager serve --host 127.0.0.1 --port 8080

An MCP client can then connect to:

http://127.0.0.1:8080/mcp

The doctor command starts the selected backend, performs MCP initialization, and lists the tools that the gateway would expose.

Agent-harness integration

Agent frameworks should depend on AgentToolProvider, not the runtime or gateway implementation:

from mcp_manager.harness import GatewayMCPClient

async with GatewayMCPClient("http://127.0.0.1:8080/mcp") as tools:
    definitions = await tools.list_tools()
    result = await tools.call_tool(
        "arxiv__search_papers",
        {"query": "mixture of experts", "max_results": 3},
    )

The model API key belongs in the agent process. Service credentials belong in backend MCP subprocess environments and are resolved through secretRef.

Current scope

  • One stdio backend: Arxiv.
  • Tool discovery and calls only.
  • One static profile.
  • Environment-backed secret provider interface, unused by Arxiv.
  • No persistence, authentication, admin API, hot reload, or distributed state.

These omissions are deliberate so the repository remains a clear starting point rather than a prematurely complex platform.

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
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
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
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