Name: EmoVideoAna
Author: hbidept

LABSCO SUMMARY

Two ways in, one account to open first.

Whichever transport you pick, the gating step is the same: an app key and secret issued by the Lianxin open platform, so budget the account signup before the install. The hosted endpoint is the shorter route if you only want to try it, because it needs no local Python at all.

What it is

A Python MCP server that fronts the Lianxin open platform (open.lianxinyun.com) for video analysis, connectable either as a local stdio process or through the vendor's hosted streamable-HTTP endpoint.

What you get

A local stdio path: the client launches the project with uv and passes your credentials in the env block
A hosted path: point the client at https://open.lianxinyun.com/mcp as a streamableHttp server, with your key pair in the sk header
Separate setup paths for Linux/macOS and for Windows, each spelled out step by step

Requirements

Generate an app key and secret at https://open.lianxinyun.com/ and pass them as APP_KEY and APP_SECRET. Locally you need uv with Python 3.13 and the mcp CLI package installed into the project. Keep the server and the chat client on the same operating system — the author warns against splitting them across Windows and WSL2 on one machine.

Setup effort

Build from source — git clone https://github.com/hbidept/emo-video-analysis-mcp.git

See installation options

USE CASES

01

Trying it without installing anything

Register the hosted streamable-HTTP endpoint in your client and put the key pair in the header.

02

Running it beside your own code

Point the client at the project directory with uv so the server starts as a local stdio process.

03

Fixing a server that never connects

Check that the client and the server are on the same operating system rather than split across Windows and WSL2.

BEFORE YOU INSTALL

What to know first

Requires

APP_KEY in your environment
APP_SECRET in your environment

Where it runs

Of the 2 documented setups, 1 runs the server on your machine and 1 connects to a hosted endpoint.

INSTALLATION

Connect your client

You clone this one yourself.

Server setup

TERMINAL

curl -LsSf https://astral.sh/uv/install.sh | sh
git clone https://github.com/hbidept/emo-video-analysis-mcp.git
cd emo-video-analysis-mcp
uv python install 3.13
uv add "mcp[cli]" httpx

MCP Clients

Any MCP client — stdio

ANY MCP CLIENT — STDIO · JSON

{
  "mcpServers": {
    "hbidept": {
      "command": "uv",
      "args": [
        "--directory",
        "<YOUR_PATH>/emo-video-analysis-mcp",
        "run",
        "main.py"
      ],
      "env": {
        "APP_KEY": "<YOUR_APP_KEY>",
        "APP_SECRET": "<YOUR_APP_SECRET>"
      }
    }
  }
}

Any MCP client — HTTP

ANY MCP CLIENT — HTTP · JSON

{
  "mcpServers": {
    "lianxin-mcp": {
      "url": "https://open.lianxinyun.com/mcp",
      "headers": {
        "sk": "appKey_sk_appSecret"
      }
    }
  }
}

EmoVideoAna