Supabase MCP Server
@MCP-Mirror
About Supabase MCP Server
Mirror of
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"alexander-zuev_supabase-mcp-server": {
"command": "uv",
"args": [
"venv"
]
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is Supabase MCP Server?
Supabase MCP Server is an MCP (Model Context Protocol) implementation that enables Cursor and Windsurf IDEs to interact directly with a Supabase PostgreSQL database. It provides a set of database management tools, including schema exploration and SQL query validation, designed for developers using these AI‑assisted IDEs.
How to use Supabase MCP Server?
Clone the repository, create a virtual environment with uv venv, install dependencies with uv sync, then configure the IDE to run the server via uv --directory <path> run main.py. For local development the server connects to 127.0.0.1:54322 with password postgres. For production projects set environment variables SUPABASE_PROJECT_REF and SUPABASE_DB_PASSWORD.
Key features of Supabase MCP Server
- Works with both Windsurf and Cursor IDEs.
- Supports local Supabase projects and production Supabase projects.
- Built-in database exploration tools with schema insights.
- Secure read-only database access.
- SQL query validation.
- Python 3.12+, PostgreSQL 16+, and uv package manager required.
Use cases of Supabase MCP Server
- Explore database schemas and tables from within Cursor or Windsurf without leaving the editor.
- Run validated SQL queries against a production Supabase database in a read‑only mode.
- Debug or inspect local Supabase PostgreSQL databases during development.
- Enable AI assistants to understand and interact with the database structure securely.
FAQ from Supabase MCP Server
What prerequisites are needed to run the server?
Python 3.12+, PostgreSQL 16+, and the uv package manager are required. On macOS, PostgreSQL must be installed before installing dependencies for the psycopg2 package.
Does the server support production Supabase databases?
Yes. Set the environment variables SUPABASE_PROJECT_REF and SUPABASE_DB_PASSWORD to connect to staging or production Supabase projects. The server remains read‑only.
Is database access read‑only?
Yes. The server enforces read‑only access to the PostgreSQL database. SQL queries are validated before execution.
How do I configure the server for Cursor or Windsurf?
For Cursor, add an MCP server entry with command: uv --directory /path/to/repo run main.py and the stdio transport. For Windsurf, edit mcp_config.json to specify the uv command, repository path, and environment variables for connection details.
What are the default connection settings for local development?
The server defaults to host 127.0.0.1:54322 with password postgres. These can be overridden for production by setting the environment variables.
Frequently asked questions
What prerequisites are needed to run the server?
Python 3.12+, PostgreSQL 16+, and the `uv` package manager are required. On macOS, PostgreSQL must be installed before installing dependencies for the `psycopg2` package.
Does the server support production Supabase databases?
Yes. Set the environment variables `SUPABASE_PROJECT_REF` and `SUPABASE_DB_PASSWORD` to connect to staging or production Supabase projects. The server remains read‑only.
Is database access read‑only?
Yes. The server enforces read‑only access to the PostgreSQL database. SQL queries are validated before execution.
How do I configure the server for Cursor or Windsurf?
For Cursor, add an MCP server entry with `command: uv --directory /path/to/repo run main.py` and the stdio transport. For Windsurf, edit `mcp_config.json` to specify the `uv` command, repository path, and environment variables for connection details.
What are the default connection settings for local development?
The server defaults to host `127.0.0.1:54322` with password `postgres`. These can be overridden for production by setting the environment variables.
Basic information
More Cloud & Infrastructure MCP servers
GCP MCP
eniayomiA Model Context Protocol (MCP) server that enables AI assistants like Claude to interact with your Google Cloud Platform environment. This allows for natural language querying and management of your GCP resources during conversations.
Supabase MCP Server
coleam00Supabase MCP server created in Python.
aws-finops-mcp-server
ravikiranvmAn MCP (Model Context Protocol) server that brings powerful AWS FinOps capabilities directly into your AI assistant. Analyze cloud costs, audit for waste, and get budget insights using natural language, all while keeping your credentials secure on your local machine.

Nippy
NippyNippy is drag-and-drop hosting (nippy.host) with an MCP server: connect it and your assistant can publish HTML, folders of photos, or file downloads to a live *.nippy.site URL — and keep editing the site in place at the
AWS MCP Servers
awslabsOpen source MCP Servers for AWS
Comments