dify-mcp-client

mcp client as agent strategy plugin.

[!important] dify is not mcp server but mcp host.

how it works

each mcp client (react agent) node can connect mcp servers.

1. tool, resource, prompt lists are converted into dify tools.
2. your selected llm can see their name, description, argument type
3. the llm calls tools based on the react loop (reason → act → observe).

[!note] most of the code in this repository contains the following files.

dify official plugins / agent strategies

https://github.com/langgenius/dify-official-plugins/tree/main/agent-strategies/cot_agent

✅ what i did

• copied react.py and renamed file as mcpreact.py
• added config_json gui input field by editing mcpreact.yaml and class mcpreactparams()

in mcpreact.py, i added

• new 12 functions for mcp
• __init__() for initializing asyncexitstack and event loop
• some codes in _handle_invoke_action() for mcp
• mcp setup and cleanup in _invoke()

[!important] react while loop is as they are

🔄 update history

• add sse mcp client (v0.0.2)
• support multi sse servers (v0.0.3)
• update python module and simplify its dependency (v0.0.4)
  • mcp(v1.1.2→v1.6.0+)
  • dify_plugin(0.0.1b72→v0.1.0)

⚠️ caution and limitation

[!caution] this plugin does not implement a human-in-the-loop mechanism by default, so connect reliable mcp server only. to avoid it, decrease max itereations(default:3) to 1, and use this agent node repeatedly in chatflow. however, agent memory is reset by the end of workflow. use conversaton variable to save history and pass it to query.
don't forget to add a phrase such as "ask for user's permission when calling tools" in instruction.

how to use this plugin

🛜install the plugin from github

• enter the following github repository name

https://github.com/3dify-project/dify-mcp-client/

• dify > plugins > + install plugin > install from > github

⬇️install the plugin from .difypkg file

• go to releases https://github.com/3dify-project/dify-mcp-client/releases
• select suitable version of .difypkg
• dify > plugins > + install plugin > install from > local package file

how to handle errors when installing plugins?

issue: if you encounter the error message: plugin verification has been enabled, and the plugin you want to install has a bad signature, how to handle the issue? solution: open /docker/.env and change from true to false:

force_verifying_signature=false

run the following commands to restart the dify service:

cd docker
docker compose down
docker compose up -d

once this field is added, the dify platform will allow the installation of all plugins that are not listed (and thus not verified) in the dify marketplace.

[!tip] marketplace need approval. if stars⭐ reach 100, i'll consider to make pr for them.

where does this plugin show up?

• it takes few minutes to install
• once installed, you can use it any workflows as agent node
• select "mcpreact" strategy (otherwise no mcp)

config

mcp agent plugin node require config_json like this to command or url to connect mcp servers

{
    "mcpservers":{
        "name_of_server1":{
            "url": "http://host.docker.internal:8080/sse"
        },
        "name_of_server2":{
            "url": "http://host.docker.internal:8008/sse"
        }
    }
}

[!warning]

• each server's port number should be different, like 8080, 8008, ...
• if you want to use stdio mcp server, there are 3 ways.
  1. convert it to sse mcp server https://github.com/3dify-project/dify-mcp-client/edit/main/readme.md#how-to-convert-stdio-mcp-server-into-sse-mcp-server
  2. deploy with source code (not by .difypkg or github reposity name install) https://github.com/3dify-project/dify-mcp-client/edit/main/readme.md#-how-to-develop-and-deploy-plugin
  3. pre-install node.js inside dify-plugin docker (issue:https://github.com/3dify-project/dify-mcp-client/issues/10) guide: https://github.com/tangyoha/tangyoha-bili/tree/master/dify/mcp/map_mcp

chatflow example

[!warning]

• the tools field should not be left blank. so select dify tools like "current time".

i provide this dify chatflow .yml for testing this plugin.

https://github.com/3dify-project/dify-mcp-client/tree/main/test/chatflow

after download dsl(yml) file, import it in dify and you can test mcp using "everything mcp server"

https://github.com/modelcontextprotocol/servers/tree/main/src/everything

how to convert stdio mcp server into sse mcp server

option1️⃣: edit mcp server's code

if fastmcp server, change like this

if __name__ == "__main__":
-    mcp.run(transport="stdio")
+    mcp.run(transport="sse")

option2️⃣: via mcp-proxy

\mcp-proxy>uv venv -p 3.12
.venv\scripts\activate
uv tool install mcp-proxy

check node.js has installed and npx(.cmd) path

(mac/linux)

which npx

(windows)

where npx

result

c:\program files\nodejs\npx
c:\program files\nodejs\npx.cmd
c:\users\user_name\appdata\roaming\npm\npx
c:\users\user_name\appdata\roaming\npm\npx.cmd

if claude_desktop_config.json is following schema,

{
  "mcpservers": {
    "server_name": {
       "command": cmd_name_or_path 
       "args": {value1, value2}
    }
  }
}

wake up stdio mcp server by this command

mcp-proxy --sse-port=8080 --pass-environment -- cmd_name_or_path --arg1 value1 --arg2 value2 ...

if your os is windows, use npx.cmd instead of npx. following is example command to convert stdio "everything mcp server" to sse via mcp-proxy.

mcp-proxy --sse-port=8080 --pass-environment -- c:\program files\nodejs\npx.cmd --arg1 -y --arg2 @modelcontextprotocol/server-everything

similarly, on another command line (if you use sample chatflow for v0.0.3)

pip install mcp-simple-arxiv
mcp-proxy --sse-port=8008 --pass-environment -- c:\users\user_name\appdata\local\programs\python\python310\python.exe -m -mcp_simple_arxiv

following is a mcp-proxy setup log.

(mcp_proxy) c:\user\user_name\mcp-proxy>mcp-proxy --sse-port=8080 --pass-environment -- c:\program files\nodejs\npx.cmd --arg1 -y --arg2 @modelcontextprotocol/server-everything
debug:root:starting stdio client and sse server
debug:asyncio:using proactor: iocpproactor
debug:mcp.server.lowlevel.server:initializing server 'example-servers/everything'
debug:mcp.server.sse:sseservertransport initialized with endpoint: /messages/
info:     started server process [53104]
info:     waiting for application startup.
info:     application startup complete.
info:     uvicorn running on http://127.0.0.1:8080 (press ctrl+c to quit)

🔨 how to develop and deploy plugin

official plugin dev guide

https://github.com/3dify-project/dify-mcp-client/blob/main/guide.md

dify plugin sdk daemon

if your os is windows and cpu is intel or amd, you need to download dify-plugin-windows-amd64.exe (v0.0.7) choose your os-compatible verson here: https://github.com/langgenius/dify-plugin-daemon/releases

1. rename it as dify.exe for convinence
2. mkdir "c\user\user\.local\bin" (windows) and register it as system path.
3. copy dify.exe to under dify-mcp-client/

[!tip] following guide is helpful. https://docs.dify.ai/plugins/quick-start/develop-plugins/initialize-development-tools

reference

https://docs.dify.ai/plugins/quick-start/develop-plugins/initialize-development-tools

[!note] you can skip this stage if you pull or download codes of this repo

dify plugin init

initial settings are as follow

change directory

cd dify-mcp-client

install python module

python3.12+ is compatible. the venv and uv are not necessary, but recommended.

uv venv -p 3.12
.venv\scripts\activate

install python modules for plugin development

uv pip install -r requirements.txt

duplicate env.example and rename one to .env

i changed remote_install_host from debug.dify.ai to localhost (docker compose environment) click 🪲bug icon button to see these information

activate dify plugin

python -m main

(ctrl+c to stop)

[!tip] remote_install_key of .env often changes. if you encounter error messages like handshake failed, invalid key, renew it.

package into .difypkg

./dify-mcp-client is my default root name

dify plugin package ./root_of_your_project

useful github repositories for developers

dify plugin sdks

https://github.com/langgenius/dify-plugin-sdks

mcp python sdk

https://github.com/modelcontextprotocol/python-sdk

[!tip] mcp client example https://github.com/modelcontextprotocol/python-sdk/blob/main/examples/clients/simple-chatbot/mcp_simple_chatbot/main.py

[!note] dify plugin has requirements.txt which automatically installs python modules. i include latest mcp in it, so you don't need to download the mcp sdk separately.