
mcp server
Refunc Model Context Protocol Server
Repository Info
About This Server
Refunc Model Context Protocol Server
Model Context Protocol (MCP) - This server can be integrated with AI applications to provide additional context and capabilities, enabling enhanced AI interactions and functionality.
Documentation
Refunc MCP Server
Deploy Refunc
Deploy all refunc components in namespace refunc-system.
kubectl apply -f https://raw.githubusercontent.com/refunc/mcp-server/refs/heads/main/deploy/play-all-in-one.yaml
Then proxy refunc service to local.
kubectl port-forward svc/aws-api 8000:http --address 0.0.0.0 -n refunc-system
kubectl port-forward svc/mcp-server 8001:http --address 0.0.0.0 -n refunc-system
PS: The data from the demo deployment is not persistent.
Create Function
Create a Function with refunc-cli.
pip install -U refunc-cli
mkdir mcp-demo && cd mcp-demo
rfctl init
Initing function manifest in /path/mcp-demo/lambda.yaml
Name: mcp-demo
Namespace: refunc-system
Select Language:
1 - python
2 - go
Choose from 1, 2 [1]:
Select Runtime:
1 - python3.10
2 - python3.9
3 - python3.8
Choose from 1, 2, 3 [1]:
# write any code in main.py, the demo is echo a hello msg.
AWS_DEFAULT_ENDPOINT=http://127.0.0.1:8000 rfctl create
PS: You need a Python 3 environment.
Create MCP Endpoint
Edit lambda.yaml as below code.
metadata:
name: mcp-demo
namespace: refunc-system
spec:
build:
source: .
manifest: requirements.txt
language: python
architecture: x86_64
handler: main.lambda_handler
timeout: 120
runtime: "python3.10"
concurrency: 1
environment:
ENV_TEST: TEST
# url:
# cors:
# allowCredentials: true
# allowHeaders: "*"
# allowMethods: "*"
# allowOrigins: "*"
# exposeHeaders: "*"
# maxAge: 300
events:
# - name: hourly
# type: cron
# mapping:
# cron: 0 * * * *
# location: Asia/Shanghai
# args:
# var1: value1
# saveLog: false
# saveResult: false
- name: mcp
type: mcp
mapping:
args:
token: mcp-demo
tools:
- name: echo-hello
desc: Echo a hello msg #Cannot contain single quotes
schema:
type: object
properties: {}
required: []
saveLog: false
saveResult: false
AWS_DEFAULT_ENDPOINT=http://127.0.0.1:8000 rfctl update-config
The demo MCP SSE endpoint is: http://127.0.0.1:8001/refunc-system/mcp-demo/test/mcp-demo/sse
Refunc MCP SSE url path format is: /namespace/<token-secret-name>/<token>/<func-name>/sse
MCP Event Spec
- name: mcp
type: mcp # the event type must be mcp.
mapping:
args:
token: mcp-demo # the token secret name you can find in play-all-in-one.yaml.
tools: # mcp tools
- name: echo-hello # mcp tool name
desc: Echo a hello msg # mcp tool desc
schema: # mcp tool parameters, described with a valid JSON Schema.
type: object
properties: {}
required: []
saveLog: false
saveResult: false
The tools field of the event is a data structure, and you can implement multiple tools using a function. When calling the function, refunc will add two built-in parameters: _call_type and _call_method, where _call_method is the name of the tool.
Quick Start
Clone the repository
git clone https://github.com/refunc/mcp-serverInstall dependencies
cd mcp-server
npm installFollow the documentation
Check the repository's README.md file for specific installation and usage instructions.
Repository Details
Recommended MCP Servers
Discord MCP
Enable AI assistants to seamlessly interact with Discord servers, channels, and messages.
Knit MCP
Connect AI agents to 200+ SaaS applications and automate workflows.
Apify MCP Server
Deploy and interact with Apify actors for web scraping and data extraction.
BrowserStack MCP
BrowserStack MCP Server for automated testing across multiple browsers.
Zapier MCP
A Zapier server that provides automation capabilities for various apps.