gRPC Transport for MCP
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@gRPC Transport for MCPcall my_tool with x=42"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
gRPC Transport for MCP
Experimental / Proof of Concept
This package is a proof of concept for experimentation purposes. It is not intended for production use. The official Python package is being developed at GoogleCloudPlatform/mcp-grpc-transport-py — please follow/star that repo to stay updated.
A gRPC transport implementation for the Model Context Protocol (MCP) using FastMCP.
Background
SEP: gRPC as a native MCP transport — the specification enhancement proposal
Related MCP server: fal
Features
Server: expose a FastMCP server over gRPC
Client: call a gRPC MCP server and get back native
mcp.typesobjectsTool calls with text and image content, structured output, and progress reporting
Proto definitions sourced from GoogleCloudPlatform/mcp-grpc-transport-proto
Installation
uv syncThe proto package (mcp-transport-proto) is fetched directly from GitHub via [tool.uv.sources] — no manual proto compilation needed.
Server
import asyncio
from mcp.server.fastmcp import FastMCP
from grpcmcp import serve_grpc
mcp = FastMCP("My Server")
@mcp.tool()
async def my_tool(x: int) -> str:
"""Does something useful."""
return str(x)
if __name__ == "__main__":
asyncio.run(serve_grpc(mcp)) # default: 0.0.0.0:50051
# asyncio.run(serve_grpc(mcp, port=9090))Client
import asyncio
from grpcmcp import GRPCClient
async def main():
async with GRPCClient("localhost", 50051) as client:
tools = await client.list_tools()
for tool in tools.tools:
print(f"{tool.name}: {tool.description}")
result = await client.call_tool("my_tool", {"x": 42})
for content in result.content:
if content.type == "text":
print(content.text)
asyncio.run(main())GRPCClient reference
Parameter | Type | Description |
|
| Server hostname |
|
| Server port |
|
| Default per-call timeout in seconds |
|
| Raw gRPC channel options |
Method | Returns |
|
|
|
|
Running the examples
Start the server:
uv run python example/grpc_example_server.pyIn another terminal, run the client:
uv run python example/grpc_example_client.pyRequirements
Python >= 3.10
grpcio >= 1.78.0
protobuf >= 4.25.0
mcp == 1.25.0
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP server for progressive tool usage at any scale (see https://klavis.ai)
MCP server for building and testing AI agents with multi-model experimentation and insights.
MCP server for secureFlows: token-free URL builders and integration-linting tools for AI agents.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceFastMCP is a comprehensive MCP server allowing secure and standardized data and functionality exposure to LLM applications, offering resources, tools, and prompt management for efficient LLM interactions.3MIT
- AlicenseAqualityDmaintenanceMCP server for interacting with fal.ai models and services. Uses the latest streaming MCP support.82MIT
- AlicenseNot gradedqualityDmaintenanceEnables MCP clients to connect to LLM/API services using the Model Context Protocol, providing real-time interaction and tool access. Also offers RESTful API endpoints via FastAPI for programmatic integration with existing systems.Apache 2.0
- FlicenseNot gradedqualityDmaintenanceSets up a Model Context Protocol (MCP) server with FastMCP, allowing LLMs to use prompts, tools, and resources.-