About MCP Argo Server
MCP Argo Server is a Go CLI that lets users run Argo Workflows through MCP. It provides tools to submit workflows, monitor their status, and retrieve results from Kubernetes and Argo resources.
Key features
- 01Submit Argo Workflows
- 02Check workflow status
- 03Wait for workflow completion
- 04Retrieve workflow results
Use cases
- 01Launch a workflow through an MCP client
- 02Monitor a submitted workflow until it finishes
- 03Fetch results from completed workflows
MCP support
The server exposes workflow submission, status, completion monitoring, and result retrieval as MCP tools. It communicates over JSON-RPC on STDIN/STDOUT.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEMCP Argo Server
An MCP-compliant server for running Argo Workflows written in Golang.
Overview
MCP Argo Server is a lightweight CLI tool that wraps Argo Workflows using JSON-RPC over STDIN/STDOUT. It leverages Foxy Contexts for RPC handling and client-go for interacting with Kubernetes and Argo Workflow resources. The project provides tools for launching workflows, checking workflow status, and retrieving results.
Installation
This project is configured to run inside a development container. Simply open the repository in your dev container-enabled editor (e.g., VS Code Remote - Containers) and all dependencies are pre-installed.
If you prefer to run it locally, clone the repository and run:
go mod tidy
Usage
Open the project in the dev container.
Run make cluster which will install the k3d cluster and set up Argo.
You can check that's worked by typing kubectl cluster-info.
You can run a test workflow by typing argo submit -n argo --watch ./kube/argo-hello-world.yaml.
You can see the Argo interface at https://localhost:2746/workflows/argo/
You can check that the app is building and the MCP is working by typing make run.
Testing with Python
The project includes a Python test client that demonstrates how to interact with the MCP Argo server. The test client is located in python/test_with_autogen.py and showcases:
- Submitting Argo workflows
- Checking workflow status
- Waiting for workflow completion
- Retrieving workflow results
To run the Python test:
-
Ensure you have Python dependencies installed:
cd python make install -
Run the test script:
python test_with_autogen.py
or... just debug it to step through.
The script will:
- Connect to the MCP Argo server
- Iterate the tools and print them out
- Submit a sample workflow from
kube/argo-hello-world.yaml - Monitor the workflow status until completion
- Display the workflow results
Contributing
Contributions are welcome! Please open issues and submit pull requests. Before submitting changes, ensure that you follow the project's coding guidelines and that all tests pass.
License
This project is licensed under the MIT License. See the LICENSE file for details.
Additional Resources
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.