xo
MCP server from sharmasuraj0123/xo-mcp-server
claude mcp add --transport stdio sharmasuraj0123-xo-mcp-server uvx git+https://github.com/sharmasuraj0123/xo-mcp-server.git \ --env ACCESS_TOKEN="your_access_token" \ --env DEPLOYMENT_ID="your_deployment_id"
How to use
This XO MCP Server provides integration with the XO platform to deploy, manage, and monitor applications directly from your MCP client. It exposes two main tool domains: Container Management and Knowledge Base Operations. Under Container Management, you can deploy applications to XO, start and stop deployed apps, remove apps, stream real-time logs, and expose apps to the internet with automatic domain provisioning. Under Knowledge Base Operations, you can update or add knowledge base content from text and ask questions to retrieve intelligent, context-aware responses for your projects. The server is built on FastMCP and relies on environment-based configuration for security. To use it, configure your MCP client to point to this server (xo-mcp-server) and provide your ACCESS_TOKEN and DEPLOYMENT_ID as environment variables. Once configured, you can issue natural language commands like deploying to XO, checking logs, or updating knowledge bases, and the server will route those commands to the appropriate XO APIs through the MCP framework.
How to install
Prerequisites:
- Python 3.11 or higher
- The XO package manager (uv) installed and available in your PATH
- An XO Launchpad account and a valid DEPLOYMENT_ID
- An XO access token for API access
Installation steps:
- Ensure Python 3.11+ is installed and available in your PATH.
- Install the XO package manager (uv). You can install it via pip:
python3 -m pip install uv
- Verify the uvx command is accessible in your shell. If needed, install the CLI wrapper or ensure your PATH includes the directory where uvx is installed.
- Install or run the XO MCP Server via MCP client configuration (no local server binary to install beyond uv). See the usage section for configuring mcpServers in your MCP client.
- Alternatively, you can test directly by cloning the repo and using the uvx runner as described in the README configuration example.
Prerequisites recap:
- Python 3.11+
- uv (XO package manager) installed
- An XO DEPLOYMENT_ID and ACCESS_TOKEN for authentication
Additional notes
Tips and common issues:
- Ensure ACCESS_TOKEN and DEPLOYMENT_ID are kept secure and not shared publicly.
- If you encounter authentication errors, re-check your DEPLOYMENT_ID in XO Launchpad and regenerate a fresh ACCESS_TOKEN if needed.
- When deploying containers, verify that your Docker image is compatible with XO and uses proper platforms (linux/amd64, linux/arm64) if you intend multi-arch support.
- For logs, make sure your MCP client is configured to stream or fetch logs from the XO deployment; the server supports real-time log retrieval.
- The configuration is environment-based; consider using a dedicated environment file or secret management in production.
- If you need to upgrade, pull the latest version from the GitHub repository and re-run the MCP client configuration with the updated server settings.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP