다른 MCP(Model Context Protocol) 서버에 연결하고 관리하는 허브 서버입니다.
이 프로젝트는 다른 MCP 서버에 연결하여 해당 서버의 도구를 나열하고 실행할 수 있는 MCP 허브 서버를 구축합니다. 특히 Cursor의 40개 도구 제한을 우회하는 데 유용합니다. Cursor 외부에서도 자주 사용되지 않는 도구를 숨겨 AI의 실수를 줄이는 데 도움이 됩니다.
Related MCP server: MCP Hub MCP Server
구성 파일을 통한 다른 MCP 서버에 대한 자동 연결
연결된 서버에서 사용 가능한 도구 나열
연결된 서버에서 도구를 호출하고 결과를 반환합니다.
mcp.json 에 다음을 추가하세요:
지엑스피1
npx 로컬 변경 사항을 적용하지 못하거나 빌드된 서버를 직접 실행하려는 경우 다음과 같이 구성할 수 있습니다.
{
"mcpServers": {
"mcp-hub": {
"command": "node",
"args": [
"/path/to/your/mcp-hub-mcp/dist/index.js",
"--config-path",
"/path/to/your/mcp-hub-config.json"
]
}
}
}참고: dist/index.js 존재하도록 프로젝트를 빌드했는지 확인하세요(예: npm run build 또는 pnpm build ).
Before processing a user's request, you must use the "list_all_tools" command to identify which tools are available.이를 통해 AI 비서가 도구를 사용하기 전에 항상 사용 가능한 도구를 확인하게 됩니다.
Node.js 18.0.0 이상
npm, yarn 또는 pnpm
# Clone repository
git clone <repository-url>
cd mcp-hub-mcp
# Install dependencies
npm install
# or
yarn install
# or
pnpm installnpm run build
# or
yarn build
# or
pnpm buildnpm start
# or
yarn start
# or
pnpm startnpm run dev
# or
yarn dev
# or
pnpm devMCP-Hub-MCP 서버는 Claude Desktop 형식의 구성 파일을 사용하여 다른 MCP 서버에 자동으로 연결합니다. 다음과 같은 방법으로 구성 파일을 지정할 수 있습니다.
환경 변수: MCP_CONFIG_PATH 환경 변수를 구성 파일 경로로 설정합니다.
명령줄 인수: --config-path 옵션을 사용하여 구성 파일 경로를 지정합니다.
기본 경로: 현재 디렉토리의 mcp-config.json 파일을 사용합니다.
구성 파일 형식:
{
"mcpServers": {
"serverName1": {
"command": "command",
"args": ["arg1", "arg2", ...],
"env": { "ENV_VAR1": "value1", ... }
},
"serverName2": {
"command": "anotherCommand",
"args": ["arg1", "arg2", ...]
}
}
}예:
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-filesystem",
"/Users/username/Desktop",
"/Users/username/Downloads"
]
},
"other-server": {
"command": "node",
"args": ["path/to/other-mcp-server.js"]
}
}
}MCP-Hub-MCP 서버는 다음과 같은 도구를 제공합니다.
list-all-tools연결된 모든 서버에서 도구 목록을 반환합니다.
{
"name": "list-all-tools",
"arguments": {}
}call-tool특정 서버에서 도구를 호출합니다.
serverName : 도구를 호출할 MCP 서버의 이름
toolName : 호출할 도구의 이름
toolArgs : 도구에 전달할 인수
{
"name": "call-tool",
"arguments": {
"serverName": "filesystem",
"toolName": "readFile",
"toolArgs": {
"path": "/Users/username/Desktop/example.txt"
}
}
}이 프로젝트는 자동 버전 관리 및 CHANGELOG 생성을 위해 기존 커밋을 따릅니다.
형식: <type>(<scope>): <description>
예:
feat: add new hub connection feature
fix: resolve issue with server timeout
docs: update API documentation
chore: update dependencies
유형:
feat : 새로운 기능(MINOR 버전 범프)
fix : 버그 수정(패치 버전 범프)
docs : 문서만 변경됨
style : 코드의 의미에 영향을 미치지 않는 변경 사항
refactor : 버그를 수정하지도 않고 기능을 추가하지도 않는 코드 변경
perf : 성능을 향상시키는 코드 변경
test : 누락된 테스트 추가 또는 기존 테스트 수정
chore : 빌드 프로세스 또는 보조 도구의 변경
주요 변경 사항: 커밋 바닥글에 BREAKING CHANGE: 추가하여 주요 버전을 올립니다.
노드.js
타입스크립트
@modelcontextprotocol/sdk (버전: ^1.9.0)
zod(버전: ^3.22.4)
MIT
Call a specific tool from a specific server
| Name | Required | Description | Default |
|---|---|---|---|
| serverName | Yes | Name of the MCP server to call tool from | |
| toolName | Yes | Name of the tool to call | |
| toolArgs | Yes | Arguments to pass to the tool |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden but only states the basic action. It lacks details on behavioral traits such as error handling, permissions needed, rate limits, or what happens when calling tools (e.g., side effects, response format). This is inadequate for a tool that interacts with other servers.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, clear sentence with zero wasted words. It's front-loaded and efficiently conveys the core purpose without unnecessary elaboration, making it easy for an agent to parse quickly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's complexity (calling tools across servers with nested parameters) and lack of annotations or output schema, the description is insufficient. It doesn't address critical context like how toolArgs should be structured, error cases, or what the call returns, leaving significant gaps for agent understanding.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema fully documents parameters (serverName, toolName, toolArgs). The description adds no additional meaning beyond the schema, such as examples or constraints, but the high coverage justifies the baseline score of 3.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Call') and target ('a specific tool from a specific server'), which is specific and unambiguous. However, it doesn't differentiate from its sibling 'list-all-tools', which serves a completely different purpose (listing vs. calling), so it misses full sibling distinction.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus alternatives. The description merely states what it does without indicating context, prerequisites, or comparison to the sibling tool 'list-all-tools', leaving the agent to infer usage scenarios.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all available tools from all connected servers. Before starting any task based on the user's request, always begin by using this tool to get a list of any additional tools that may be available for use.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It describes the tool's behavior as listing tools from all servers and suggests it should be used first in tasks, which adds useful context. However, it lacks details on potential limitations (e.g., rate limits, authentication needs, or what 'all connected servers' entails), leaving some behavioral aspects unclear.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise and well-structured in two sentences: the first states the purpose, and the second provides usage guidelines. Every sentence adds value without redundancy. It could be slightly more front-loaded by emphasizing the 'list' action first, but overall it's efficient and clear.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (0 parameters, no output schema, no annotations), the description is reasonably complete. It explains what the tool does and when to use it, which suffices for this context. However, it lacks details on output format or behavioral constraints (e.g., whether it's idempotent or has side effects), preventing a perfect score.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 0 parameters with 100% coverage, so no parameter documentation is needed. The description appropriately does not discuss parameters, focusing instead on the tool's purpose and usage. This meets the baseline of 4 for tools with no parameters, as it avoids unnecessary details.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'List all available tools from all connected servers.' It specifies the verb ('list') and resource ('tools'), and distinguishes it from the sibling tool 'call-tool' by focusing on enumeration rather than invocation. However, it doesn't explicitly differentiate from potential other listing tools beyond the sibling, keeping it at 4 instead of 5.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides explicit usage guidelines: 'Before starting any task based on the user's request, always begin by using this tool to get a list of any additional tools that may be available for use.' It clearly states when to use it (at the start of tasks) and why (to discover additional tools), offering strong guidance without alternatives needed given the context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
call-toollist-all-toolsScored across 2 tools
The two tools have completely distinct purposes with no overlap: one is for calling specific tools from servers, while the other is for listing all available tools. An agent can easily tell them apart as they serve different functions in the tool management workflow.
The tool names follow a consistent verb_noun pattern (call-tool, list-all-tools) with clear actions and objects. The hyphenated style is used consistently, though 'list-all-tools' includes a modifier ('all') which is a minor deviation from a pure verb_noun structure.
With only 2 tools, the server feels thin for its apparent purpose of managing MCP tools across servers. While it covers basic operations (call and list), more comprehensive tool management (e.g., discover servers, configure connections) is missing, making the scope underdeveloped.
The tool surface is severely incomplete for managing an MCP hub. It lacks essential operations such as discovering or connecting to servers, managing server configurations, or handling errors. The two tools provide only minimal functionality, leaving significant gaps that will hinder agent workflows.
The Remote MCP server acts as a standardized bridge between LLM applications (like Claude, ChatGPT, and Cursor) and external services, enabling AI agents to access external tools and resources. Its primary capability is providing a centralized search tool to discover other MCP servers and their respective tools. Unlike local implementations, it runs remotely with OAuth authentication and permission controls for security.
An MCP server that integrates with Discord to provide AI-powered features.
AI tool directory MCP server, live since July 2026. Tools, agents, models, companies, services.
One AI endpoint to search and call 22k+ MCP servers; 50+ hosted tools work instantly, no key.
