Xano MCP Server for Smithery
A Model Context Protocol (MCP) server for integrating Xano databases with Smithery, enabling Claude AI to interact with Xano databases.
Overview
This MCP server provides a bridge between Claude AI (via Smithery) and Xano databases, allowing Claude to perform operations on Xano data through a standardized interface. The server implements the Model Context Protocol, making it compatible with Smithery’s serverless deployment model.
Features
- Complete Xano API integration
- Support for both stdio and WebSocket transport methods
- Comprehensive database operations (tables, schemas, records)
- File management capabilities
- Request history tracking
- Import/export functionality
Available Tools
The server provides the following categories of tools:
Instance and Database Operations
- List Xano instances
- Get instance details
- List databases/workspaces
- Get workspace details
Table Operations
- List tables
- Get table details
- Create, update, and delete tables
Table Schema Operations
- Get and update table schemas
- Add, rename, and delete fields
Table Index Operations
- List, create, and delete various index types (btree, unique, search, spatial, vector)
Table Content Operations
- Browse and search table content
- CRUD operations on records (create, read, update, delete)
- Bulk operations for efficiency
File Operations
- List, upload, and delete files
- Get file details
Request History Operations
- Browse and search request history
Workspace Import/Export
- Export and import workspaces and schemas
Installation
Prerequisites
- Python 3.10 or higher
- Smithery CLI (for deployment)
- Xano API token
Local Installation
# Clone the repository
git clone https://github.com/roboulos/xano-mcp.git
cd xano-mcp
# Install dependencies
pip install -r requirements.txt
Usage
Running Locally
# Run with stdio transport (default)
python -m src.xano_mcp --token YOUR_XANO_API_TOKEN
# Run with WebSocket transport
python -m src.xano_mcp --token YOUR_XANO_API_TOKEN --transport websocket --port 8765
# Enable debug mode
python -m src.xano_mcp --token YOUR_XANO_API_TOKEN --debug
Using with Smithery
- Deploy the MCP server to Smithery:
smithery deploy
Configure the server with your Xano API token in the Smithery dashboard
Use the server in your Smithery workflows
Security Considerations
- Store your Xano API token securely
- Use environment variables for sensitive information when possible
- Consider using access controls on your Xano database
- The MCP server has full access to your Xano database, so deploy it securely
Configuration
The server can be configured using command-line arguments or environment variables:
Option | Environment Variable | Description |
---|---|---|
–token | XANO_API_TOKEN | Your Xano API token (required) |
–transport | MCP_TRANSPORT | Transport method: stdio or websocket (default: stdio) |
–port | MCP_PORT | Port for WebSocket server (default: 8765) |
–debug | MCP_DEBUG | Enable debug mode for verbose logging |
Docker Support
You can run the server using Docker:
# Build the Docker image
docker build -t xano-mcp .
# Run with stdio transport
docker run -e XANO_API_TOKEN=YOUR_TOKEN xano-mcp
# Run with WebSocket transport
docker run -e XANO_API_TOKEN=YOUR_TOKEN -p 8765:8765 xano-mcp --transport websocket --port 8765
License
This project is licensed under the MIT License - see the LICENSE file for details.
Xano MCP Server for Smithery
Project Details
- roboulos/xano-mcp
- MIT License
- Last Updated: 3/23/2025
Recomended MCP Servers
Share code with LLMs via Model Context Protocol or clipboard. Rule-based customization enables easy switching between different tasks...
A working pattern for SSE-based MCP clients and servers
Audiense Insights MCP Server is a server based on the Model Context Protocol (MCP) that allows Claude and...
Model Context Protocol server that integrates AgentQL's data extraction capabilities.
MCP Server for querying DBT Semantic Layer
A MCP (Model Context Protocol) server implementation for SimpleLocalize.
MCP server for apple notes
A Model Context Protocol Server for Interacting with Slack