Skip to main content

Remote MCP Server

A server implementation for Model Context Protocol (MCP) that enables remote access to MCP capabilities, developed by Fetcch.

Overview

Remote MCP Server extends the traditional local-only MCP implementation by providing a HTTP API interface, allowing any MCP to be accessed remotely. This enables distributed architectures where MCP-powered tools, resources, and prompts can be accessed over a network.

Features

  • Remote access to MCP tools, resources, and prompts
  • RESTful API interface
  • Easy configuration and setup
  • Compatible with existing MCP clients

Installation

# Using uv (recommended)
uv pip install remote-mcp-server

# Or using pip
pip install remote-mcp-server

Creating an MCP Server

Creating a remote MCP server is straightforward. Here's a basic example:

from remote_mcp_server.server.config import Config
from remote_mcp_server.server.main import MCPServer
from mcp.types import ServerCapabilities

# Initialize server configuration
config = Config(
    warn_on_duplicates=True, 
    capabilities=ServerCapabilities(), 
    name="MyRemoteMCP", 
    version="1.0.0"
)

# Create server instance
server = MCPServer(config=config)

# Start the server
app = server.start_server()

# Run with uvicorn
# uvicorn my_server:app --host 0.0.0.0 --port 8000

Adding Tools

Tools are functions that can be called remotely through the MCP protocol:

async def add_numbers(a: int, b: int) -> str:
    """Adds two integers and returns the result as a string."""
    return str(a + b)

server.add_tool(
    name="add",
    description="Add two numbers together",
    input_schema={
        "title": "InputSchema",
        "type": "object",
        "properties": {
            "a": {"title": "a", "type": "integer"},
            "b": {"title": "b", "type": "integer"}
        },
        "required": ["a", "b"]
    },
    fn=add_numbers
)

Adding Resources

Resources are data that can be accessed remotely:

async def read_text_resource(uri: str) -> list[ReadResourceResult]:
    """Read a text resource by URI."""
    return [
        ReadResourceResult(contents=[
            TextResourceContents(
                uri=uri, 
                mimeType="text/plain", 
                text="Resource content here"
            )
        ])
    ]

def list_text_resources() -> list[Resource]:
    """List available text resources."""
    return [
        Resource(
            uri="text:///example",
            name="ExampleText",
            description="An example text resource",
            mimeType="text/plain"
        )
    ]

resource = ResourceModel(
    uri="text:///example",
    name="ExampleText",
    description="An example text resource",
    mime_type="text/plain",
    size=123,
    list_fn=list_text_resources,
    read_fn=read_text_resource
)

server.add_resource(resource)

Adding Prompts

Prompts are template messages that can be customized with arguments:

def generate_greeting(args: Dict[str, str]) -> list[PromptMessage]:
    """Generate a greeting based on provided name."""
    name = args.get("name", "Guest")
    return [
        PromptMessage(
            role="assistant", 
            content=TextContent(
                type="text", 
                text=f"Hello, {name}! How can I help you today?"
            )
        )
    ]

greeting_prompt = PromptModel(
    name="greeting",
    description="Generates a personalized greeting",
    arguments=[
        ArgumentModel(
            name="name",
            description="The name to greet",
            required=False
        )
    ],
    prompt_fn=generate_greeting
)

server.add_prompt(greeting_prompt)

Running the Server

To run a remote MCP server:

# Save as server.py
from remote_mcp_server.server.config import Config
from remote_mcp_server.server.main import MCPServer
from mcp.types import ServerCapabilities
import uvicorn

# Initialize server with configuration
config = Config(
    warn_on_duplicates=True, 
    capabilities=ServerCapabilities(), 
    name="MyRemoteMCP", 
    version="1.0.0"
)
server = MCPServer(config=config)

# Add your tools, resources, and prompts here...

# Start the server
app = server.start_server()

if __name__ == "__main__":
    uvicorn.run(app, host="0.0.0.0", port=8000)

Then run:

python server.py

Configuring MCP Client

To connect to a remote MCP server from an MCP client, add the following to your config.json:

{
  "mcpServers": {
    "client-mcp": {
      "command": "/path/to/uv",
      "args": [
        "--directory", 
        "/path/to/mcp-servers",
        "run",
        "/path/to/mcp-servers/mcp-local-server/cli.py",
        "http://localhost:8000"  // URL of your remote MCP server
      ]
    }
  }
}

Example

A complete example server with an add tool and welcome prompt:

# examples/add_with_prompt.py
from remote_mcp_server.server.config import Config, ResourceModel, PromptModel, ArgumentModel
from mcp.types import ServerCapabilities, ReadResourceResult, TextResourceContents, Resource, PromptMessage, TextContent
from typing import Dict
import asyncio
from remote_mcp_server.server.main import MCPServer
import uvicorn

# Initialize server configuration
config = Config(
    warn_on_duplicates=True, 
    capabilities=ServerCapabilities(), 
    name="ExampleMCP", 
    version="1.0.0"
)
server = MCPServer(config=config)

# Add tool implementation
async def add(a: int, b: int) -> str:
    return str(a + b)

server.add_tool(
    "add",
    "add 2 numbers", 
    {
        "title": "InputSchema",
        "type": "object",
        "properties": {
            "a": {"title": "a", "type": "integer"},
            "b": {"title": "b", "type": "integer"}
        },
        "required": ["a", "b"]
    }, 
    add
)

# Add welcome prompt
def generate_welcome_prompt(args: Dict[str, str]) -> list[PromptMessage]:
    name = args.get("username", "Guest")
    return [PromptMessage(
        role="assistant", 
        content=TextContent(type="text", text=f"Welcome aboard, {name}!")
    )]

welcome_prompt = PromptModel(
    name="welcome_message",
    description="Generates a welcome message for users",
    arguments=[
        ArgumentModel(
            name="username",
            description="The username to welcome",
            required=False
        )
    ],
    prompt_fn=generate_welcome_prompt
)
server.add_prompt(welcome_prompt)

# Start the server
app = server.start_server()

if __name__ == "__main__":
    uvicorn.run(app, host="0.0.0.0", port=8000)

Metadata

Release files for local-mcp-server 0.2.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for local-mcp-server 0.2.2
File Size Uploaded Provenance
local_mcp_server-0.2.2.tar.gz 6.0 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for local-mcp-server 0.2.2
File Interpreter ABI Platform
local_mcp_server-0.2.2-py3-none-any.whl Python 3 none any Details

Total release size: 13.7 kB

Release files / local_mcp_server-0.2.2.tar.gz

Download URL local_mcp_server-0.2.2.tar.gz
Size 6.0 kB
Tags Source
SHA-256 checksum
How to use checksums
6395e379968fdd94b11d4c3cda1990661246094333427c37703b09e8d1608779
BLAKE2b-256 checksum
How to use checksums
8c61f9adedee41ffa48083c6bbea50d8c69a322f950a85d4802a6df1111e66e2
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.6.7

Release files / local_mcp_server-0.2.2-py3-none-any.whl

Download URL local_mcp_server-0.2.2-py3-none-any.whl
Size 7.7 kB
Tags Python 3
SHA-256 checksum
How to use checksums
1f7aeeb12c45c4a2ccac2877adc3d497094e5f3f82724e4b7f19abdaca3fe3c4
BLAKE2b-256 checksum
How to use checksums
e08ee8864f99bf480fea59bfdd43edeadf11801d2b9bfc190fca22d88d298be7
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via uv/0.6.7

Security for local-mcp-server 0.2.2

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page