TianGong AI

tiangong-ai-mcp-local

typescript

1

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "linancn-tiangong-ai-mcp-local": {
      "command": "docker",
      "args": [
        "run",
        "-d",
        "--name",
        "tiangong-ai-mcp-server-local",
        "--publish",
        "9279:9279",
        "--env-file",
        ".env",
        "linancn/tiangong-ai-mcp-server-local:0.0.1"
      ],
      "env": {
        "ENV_FILE": "YOUR_ENV_FILE_PATH"
      }
    }
  }
}

TianGong AI MCP Server provides a local, streamable HTTP protocol interface that lets your client send requests to the TianGong AI model context mechanism and receive responses in a streaming fashion. It is useful for development, testing, and integrating TianGong AI capabilities into your applications without relying on remote services.

How to use

You run the local MCP server and then connect your MCP client to it using the standard MCP HTTP workflow. The server is designed to operate locally and can be managed with common process tools. Once started, your client can stream prompts and get model-context results back in real time, enabling interactive sessions and efficient data exchange. Use the provided local runtime commands to start the server, then point your client at the local endpoint and begin sending requests.

How to install

Prerequisites you need before installation: Node.js (for npm and npx) and Docker if you plan to run via container. You should also have npm and npx available in your shell.

Install and start the MCP server using the local HTTP workflow:

npm install -g @tiangong-ai/mcp-server-local

npx dotenv -e .env -- \
npx -y -p @tiangong-ai/mcp-server-local tiangong-ai-mcp-http

npm i -g pm2

pm2 start "npx --no-install tiangong-ai-mcp-http" --name tiangong-mcp-local --time

pm2 restart tiangong-mcp-local
pm2 stop tiangong-mcp-local
pm2 logs tiangong-mcp-local

pm2 delete tiangong-mcp-local

pm2 status

Additional notes

You can also run the MCP server using Docker. This is convenient for isolating dependencies and running the server consistently across environments.

# Build MCP server image using Dockerfile (optional)
docker build -t linancn/tiangong-ai-mcp-server-local:0.0.1 .

# Pull MCP server image
docker pull linancn/tiangong-ai-mcp-server-local:0.0.1

# Start MCP server using Docker
docker run -d \
    --name tiangong-ai-mcp-server-local \
    --publish 9279:9279 \
    --env-file .env \
    linancn/tiangong-ai-mcp-server-local:0.0.1