Yunikorn
Make Yunikorn accessible to AI agents! Queues, partitions, applications, and nodes become readable to LLMs.
What is Yunikorn?
Yunikorn is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to make yunikorn accessible to ai agents! queues, partitions, applications, and nodes become readable to llms.
Make Yunikorn accessible to AI agents! Queues, partitions, applications, and nodes become readable to LLMs.
This server falls under the Cloud Services category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Make Yunikorn accessible to AI agents! Queues, partitions, a
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx yunikornConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Yunikorn
The YuniKorn MCP Server exposes Apache YuniKorn's Kubernetes batch scheduler to AI agents via the Model Context Protocol. It provides 7 MCP tools and 2 MCP resources covering partitions, queues, applications, nodes, user resource usage, and scheduler health—enabling LLMs to observe and reason about batch workload scheduling without direct cluster access.
Prerequisites
- Python 3.10 or later (3.12+ recommended)
- uv package manager installed (https://docs.astral.sh/uv/)
- Access to a running Apache YuniKorn scheduler REST API endpoint
- Git to clone the repository
- Docker (optional, for containerized deployment)
Clone the repository
Clone the yunikorn-mcp-server repository from GitHub.
git clone https://github.com/frenoid/yunikorn-mcp-server.git
cd yunikorn-mcp-serverInstall dependencies
Use uv to install the Python dependencies listed in requirements.txt. Activate a virtual environment first if you are using one.
python -m venv .venv
source .venv/bin/activate
uv pip install -r requirements.txtSet the YuniKorn base URL
Export the YUNIKORN_BASE_URL environment variable pointing to your YuniKorn REST API. The default assumes localhost on port 9089. Set TLS_INSECURE=true only if your endpoint uses a self-signed certificate.
export YUNIKORN_BASE_URL=http://yunikorn.example.com:9089/ws/v1/
export TLS_INSECURE=false # set true for self-signed certsRun the server in stdio mode for IDE / Claude Code integration
Start the server using stdio transport, which is required for Claude Desktop and Claude Code integration.
uv run python -m main --transport stdioAlternatively run as HTTP server or via Docker
For browser-based MCP Inspector access or multi-client setups, run with the streamable-http transport. The MCP endpoint is exposed at /mcp on port 8000. A Docker image is available for containerized deployment.
# HTTP mode (default)
uv run python -m main --transport streamable-http --host 0.0.0.0 --port 8000
# Docker
docker run -p 8000:8000 \
-e YUNIKORN_BASE_URL="http://yunikorn.example.com:9089/ws/v1" \
docker.io/frenoid/yunikorn-mcp-server:latestAdd to Claude Desktop configuration
Register the server in claude_desktop_config.json so Claude Desktop can spawn it automatically.
{
"mcpServers": {
"yunikorn": {
"command": "/path/to/.venv/bin/python",
"args": ["-m", "main", "--transport", "stdio"],
"cwd": "/path/to/yunikorn-mcp-server",
"env": {
"YUNIKORN_BASE_URL": "http://yunikorn.example.com:9089/ws/v1/"
}
}
}
}Yunikorn Examples
Client configuration
Claude Desktop configuration for the YuniKorn MCP server in stdio mode with the YuniKorn URL passed as an environment variable.
{
"mcpServers": {
"yunikorn": {
"command": "python",
"args": ["-m", "main", "--transport", "stdio"],
"cwd": "/home/user/yunikorn-mcp-server",
"env": {
"YUNIKORN_BASE_URL": "http://yunikorn.cluster.local:9089/ws/v1/"
}
}
}
}Prompts to try
Natural language questions you can ask once the YuniKorn MCP server is connected.
- "List all partitions in the YuniKorn scheduler"
- "Show me the queue hierarchy for the default partition"
- "Which applications are currently running in the root.spark queue?"
- "Check the overall health of the YuniKorn scheduler"
- "How much CPU and memory is each user consuming in the default partition?"
- "Show node utilization across the cluster"Troubleshooting Yunikorn
Connection refused or timeout when calling YuniKorn tools
Verify YUNIKORN_BASE_URL is correct and the YuniKorn REST API port (default 9089) is reachable from the server host. If running inside a Kubernetes cluster, ensure the service is exposed or use kubectl port-forward.
SSL certificate verification errors on HTTPS endpoints
Set the environment variable TLS_INSECURE=true when connecting to a YuniKorn instance with a self-signed or internal CA certificate. Do not use this in production environments without understanding the security implications.
Memory values appear as very large integers
YuniKorn represents memory in raw bytes (64-bit signed integers) and CPU in millicores. Divide memory values by 1073741824 to convert to GiB, and CPU millicore values by 1000 to get full vCPU cores.
Frequently Asked Questions about Yunikorn
What is Yunikorn?
Yunikorn is a Model Context Protocol (MCP) server that make yunikorn accessible to ai agents! queues, partitions, applications, and nodes become readable to llms. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Yunikorn?
Follow the installation instructions on the Yunikorn GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Yunikorn?
Yunikorn works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Yunikorn free to use?
Yes, Yunikorn is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Yunikorn Alternatives — Similar Cloud Services Servers
Looking for alternatives to Yunikorn? Here are other popular cloud services servers you can use with Claude, Cursor, and VS Code.
Open WebUI
★ 138.2kUser-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Anything LLM
★ 60.4kThe all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
LocalAI
★ 46.4kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Nacos
★ 33.0kan easy-to-use dynamic service discovery, configuration and service management platform for building AI cloud native applications.
Xiaozhi ESP32
★ 26.7k本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Gateway
★ 11.8kA blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.
Browse More Cloud Services MCP Servers
Explore all cloud services servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Yunikorn in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Yunikorn?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.