Supabase MCP Server
About
A Model Context Protocol (MCP) server that provides tools for interacting with a Supabase database, enabling AI assistants to perform database operations through a standardized interface. It is a demonstration project, work in progress.
Details
- Author
- coleam00
- GitHub stars
- 195
- Downloads
- 508
- Categories
- Cloud Service, Database
Jump to
- Read table rows with optional filtering, pagination, and column selection
- Create new records into Supabase tables
- Update existing records based on filters
- Delete records based on filters
- Runs as a Docker container for easy integration
- Uses the Model Context Protocol for AI assistant communication
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Supabase MCP ServerCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Clone the repository, build the Docker image with docker build -t mcp/supabase ., then configure your MCP client (e.g., Claude Desktop, Windsurf) with the provided JSON config, setting SUPABASE_URL and SUPABASE_SERVICE_KEY environment variables. The server runs via Docker and exposes four tools: read_table_rows, create_table_records, update_table_records, delete_table_records.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"supabase mcp server": {
"supabase-mcp-coleam00": {
"command": "docker",
"args": [
"build",
"-t",
"mcp/supabase",
"."
]
}
}
}
}McpServers
{
"supabase-mcp-coleam00": {
"command": "docker",
"args": [
"build",
"-t",
"mcp/supabase",
"."
]
}
}Supabase MCP Server
A Model Context Protocol (MCP) server that provides tools for interacting with a Supabase database. This server enables AI assistants to perform database operations through a standardized interface.
NOTE: This Supabase MCP server was created as a demonstration of my AI IDE coding workflow. It is still a work in progress which I will expand on in future videos on my channel.
Features
- Read Table Rows: Query data from Supabase tables with optional filtering, pagination, and column selection
- Create Table Records: Insert new records into Supabase tables
- Update Table Records: Modify existing records in Supabase tables based on filters
- Delete Table Records: Remove records from Supabase tables based on filters
Prerequisites
- Docker or Docker Desktop
- Supabase account and project
Installation
1. Clone the repository:
git clone https://github.com/coleam00/supabase-mcp.git
cd supabase-mcp
Docker Setup
1. Build the Docker image:
docker build -t mcp/supabase .
Usage
Running as an MCP Server with Docker
The Supabase MCP server can be integrated with AI assistants using the Model Context Protocol.
1. Include the below configuration in your MCP config (in Claude Desktop, Windsurf, etc.)
Be sure to build the container with the installation steps first!
{
"mcpServers": {
"supabase": {
"command": "docker",
"args": ["run", "--rm", "-i", "-e", "SUPABASE_URL", "-e", "SUPABASE_SERVICE_KEY", "mcp/supabase"],
"env": {
"SUPABASE_URL": "YOUR-SUPABASE-URL",
"SUPABASE_SERVICE_KEY": "YOUR-SUPABASE-SERVICE-ROLE-KEY"
}
}
}
}
2. Replace YOUR-SUPABASE-URL and YOUR-SUPABASE-SERVICE-ROLE-KEY with your actual Supabase credentials.
3. The AI assistant can now access the Supabase database through the MCP server using the provided tools.
For more information on the Model Context Protocol, visit modelcontextprotocol.io.
Available Tools
Read Table Rows
read_table_rows(
table_name: str,
columns: Optional[List[str]] = None,
filters: Optional[Dict[str, Any]] = None,
limit: Optional[int] = None,
offset: Optional[int] = None
)
Example:
```python
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
