Glide MCP Server

Model Context Protocol (MCP) server for interacting with Glide Apps API. This server enables AI assistants to communicate with Glide Apps through a standard interface.

Features

• Supports both standard Glide Tables API and Glide Big Tables API v2
• Provides comprehensive access to Glide's data management capabilities
• Big Tables support for datasets up to 10 million rows
• Stashing for large data uploads
• Pagination for handling large datasets
• Multiple deployment options: Docker, local, Claude Desktop, RooCode, Windsurf
• HTTP adapter for services that require HTTP endpoints

Quick Start

Option 1: One-Click Setup (Recommended for Beginners)

1. Clone this repository:

   git clone https://github.com/mows21/glide-mcp-server-v2.git
   cd glide-mcp-server-v2
   

2. Run the setup script:

   chmod +x setup.sh
   ./setup.sh
   

3. Follow the prompts to enter your Glide API key and configure options.

Option 2: Manual Setup with Docker Compose

1. Clone the repository:

   git clone https://github.com/mows21/glide-mcp-server-v2.git
   cd glide-mcp-server-v2
   

2. Create a .env file with your Glide API key:

   GLIDE_API_KEY=your_api_key_here
   HTTP_ADAPTER=false
   PORT=3000
   USE_BIG_TABLES=false
   

3. Run with Docker Compose:

   docker-compose up -d
   

Option 3: Docker Run

docker build -t glide-mcp-server .
docker run -it --rm --env-file .env glide-mcp-server

Available APIs

Standard Glide Tables API

The standard API works with regular Glide Tables and provides basic CRUD operations. This is suitable for tables with up to 25,000 rows.

Glide Big Tables API v2

This API is designed specifically for Glide Big Tables, supporting up to 10 million rows. It includes features like:

• Stashing for large data uploads
• Pagination for large result sets
• Lower update usage costs
• SQL query support

To use the Big Tables API, set USE_BIG_TABLES=true in your .env file or use the --big-tables flag when running the unified server.

Integration with AI Assistants

Claude Desktop

1. Make sure the MCP server is running with HTTP adapter enabled:

   HTTP_ADAPTER=true
   

2. Configure Claude Desktop to use the server:

   • Copy the claude-config.json file to your Claude Desktop configuration directory
   • Ensure the API key is properly set

RooCode

1. Install the tool with the provided configuration:

   roocode tools install --config ./roocode-config.js
   

Windsurf

1. Install the tool with the provided configuration:

   windsurf tools install --config ./windsurf.yaml
   

Available Tools

Standard Glide Tables API Tools

App Management

• list_apps: List all accessible Glide apps
• get_app_details: Get detailed information about a specific app

Table Management

• list_tables: List all tables in an app
• get_table_details: Get detailed information about a specific table
• list_columns: List all columns in a table

Data Operations

• get_rows: Retrieve rows from a table with optional filtering
• get_row: Get a specific row by ID
• create_rows: Create new rows in a table
• update_row: Update an existing row
• delete_row: Delete a row from a table

Admin API

• User and role management tools
• Application settings tools

Glide Big Tables API v2 Tools

Table Management

• list_tables: List all tables accessible to the API key
• create_table: Create a new Big Table with optional initial rows
• create_table_from_stash: Create a new Big Table using data from a stash

Data Operations

• get_rows: Retrieve rows from a table with pagination support
• get_row: Get a specific row by ID
• add_rows: Add new rows to a table
• add_rows_from_stash: Add rows to a table from a stash
• update_row: Update an existing row
• delete_row: Delete a row from a table
• overwrite_table: Overwrite all rows in a table
• overwrite_table_from_stash: Overwrite all rows in a table using data from a stash

Stash Operations

• create_stash_chunk: Create or update a chunk in a stash
• delete_stash: Delete a stash

Helper Tools

• get_all_rows: Helper to retrieve all rows from a table using pagination
• stash_large_dataset: Helper to stash a large dataset in chunks

Configuration Options

The server can be configured using environment variables in the .env file:

• GLIDE_API_KEY: Your Glide API key (required)
• HTTP_ADAPTER: Set to "true" to enable HTTP adapter (default: "false")
• PORT: Port for HTTP adapter if enabled (default: 3000)
• USE_BIG_TABLES: Set to "true" to use Glide Big Tables API v2 (default: "false")

Advanced Configuration

HTTP Adapter

The HTTP adapter allows you to use the MCP server with services that require HTTP endpoints rather than stdin/stdout communication. To enable it, set HTTP_ADAPTER=true in your .env file.

The HTTP adapter exposes the following endpoints:

• POST /mcp: The MCP endpoint that accepts MCP requests
• GET /health: A health check endpoint

Customizing Docker Deployment

You can customize the Docker deployment by modifying the docker-compose.yml file. For example, you can:

• Change the port mapping
• Add additional environment variables
• Mount custom volumes

Troubleshooting

Common Issues

• API Key Issues: Ensure your API key is correctly set in the .env file
• Connection Problems: Check that your firewall allows connections to the server
• Integration Issues: Verify that your AI assistant is correctly configured to use the MCP server
• Table Type Mismatch: Make sure you're using the correct API for your table type:
  • For regular Glide Tables, set USE_BIG_TABLES=false
  • For Big Tables, set USE_BIG_TABLES=true

License

MIT

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.