Supabase

You run a self-hosted Supabase MCP Server to manage and automate administration tasks for your private Supabase instance. It exposes a rich set of MCP tools for database, auth, storage, real-time, migrations, monitoring, and more, enabling you to perform maintenance, monitoring, and governance directly against your Supabase deployment from a single, centralized endpoint.

How to use

Use an MCP client to connect to the server via either the HTTP self-hosted endpoint or the local stdio process you run on your machine or CI environment. The server exposes a main JSON-RPC endpoint for tool execution and a discovery endpoint that lists available tools. You can perform administrative actions, run migrations, query database stats, manage users and roles, handle storage, and monitor health and performance. Start by configuring your client with your server endpoint and the required Supabase credentials, then select the tool you want to run from the available tools list and execute it through the client. All actions are secured with input validation, rate limiting, and audit logging.

How to install

Prerequisites: you need Python 3.11+ and a Supabase instance (self-hosted or cloud). You will also set environment variables for the Supabase project.

Option A — SDK Smithery (recommended) follow these steps to install and run the MCP server via Smithery dev flow.

# Clone the repository
git clone https://github.com/MisterSandFR/Supabase-MCP-SelfHosted.git
cd Supabase-MCP-SelfHosted

# Install dependencies
pip install -r requirements.txt

# (Optional) Install the Smithery CLI if needed
npm i -g @smithery/cli

# Run the dev SDK (choose the environment you have)
smithery dev   # or: smithery playground

Option B — HTTP self-hosted (compat) follow these steps to run the HTTP interface directly against your Supabase instance.

# Clone the repository
git clone https://github.com/MisterSandFR/Supabase-MCP-SelfHosted.git
cd Supabase-MCP-SelfHosted

# Install Python dependencies
pip install -r requirements.txt

# Configure environment variables
export SUPABASE_URL="https://your-project.supabase.co"
export SUPABASE_ANON_KEY="your-anon-key"
export SUPABASE_SERVICE_KEY="your-service-key"  # Optional

# Start the external HTTP server
python src/supabase_server.py

With Docker (Railway / self-hosted) you can build and run a container that hosts the MCP server.

# Build and run (use the Railway-specific Dockerfile if needed)
docker build -f Dockerfile.railway -t supabase-mcp-server .
docker run -p 8000:8000 \
  -e SUPABASE_URL="https://your-project.supabase.co" \
  -e SUPABASE_ANON_KEY="your-anon-key" \
  supabase-mcp-server

Additional configuration and credentials

Environment variables you will configure to connect to your Supabase project.

# Supabase Configuration
SUPABASE_URL=https://your-project.supabase.co
SUPABASE_ANON_KEY=eyJ... (your anon key)
SUPABASE_SERVICE_KEY=eyJ... (optional)

# Server configuration
PORT=8000
PYTHONUNBUFFERED=1

Security and monitoring notes

The server validates inputs to prevent SQL injections, applies rate limiting by IP and user, keeps audit logs of all operations, requires HTTPS in production, securely handles secrets, and enforces Row Level Security policies for data protection. Real-time metrics, logs, health checks, and performance analysis help you maintain visibility into your MCP-enabled Supabase instance.

Deployment and workflow notes

You can deploy using Smithery, Railway, or Docker. Smithery is the recommended path for Python-based MCP servers. Railways provides a quick self-hosted path, and Docker offers containerized deployment for reproducible environments.

Troubleshooting and tips

If you encounter issues, verify that your Supabase URL and keys are correct, ensure the server port 8000 is reachable, and check the audit logs for failed operations. For cached or stale profiles in Smithery, re-save the profile and rerun the Scan.

Notes on integration

This MCP server is designed to work smoothly with the MCP Hub Central integration, enabling centralized discovery and orchestration of MCP servers within your environment.

Available tools

execute_sql

Execution of SQL queries against the database

list_tables

List database tables

inspect_schema

Inspect database schema

apply_migration

Apply migrations to the database

backup_database

Backup database content

restore_database

Restore database from a backup

vacuum_analyze

Vacuum and analyze database for maintenance

get_database_stats

Query database statistics

create_index

Create database index

drop_index

Drop database index

list_extensions

List database extensions

manage_extensions

Manage database extensions

execute_psql

Run psql commands

check_health

Check database health

get_database_connections

List active database connections

list_auth_users

List authentication users

create_auth_user

Create an authentication user

update_auth_user

Update an authentication user

delete_auth_user

Delete an authentication user

get_auth_user

Get authentication user details

verify_jwt_secret

Verify JWT secret

manage_roles

Manage authentication roles

manage_rls_policies

Manage Row Level Security policies

list_storage_buckets

List storage buckets

list_storage_objects

List storage objects

manage_storage_policies

Manage storage policies

upload_file

Upload a storage file

download_file

Download a storage file

delete_file

Delete a storage file

list_realtime_publications

List real-time publications

manage_realtime

Manage real-time settings

create_subscription

Create a real-time subscription

delete_subscription

Delete a real-time subscription

create_migration

Create a migration

list_migrations

List migrations

push_migrations

Push migrations to the server

validate_migration

Validate a migration

smart_migration

Smart migration tooling

auto_migrate

Automatic migrations

sync_schema

Synchronize database schema

import_schema

Import a database schema

get_logs

Retrieve server logs

metrics_dashboard

View metrics dashboard

analyze_performance

Analyze performance across operations

analyze_rls_coverage

Analyze RLS coverage

audit_security

Audit server security

generate_typescript_types

Generate TypeScript types from schemas

generate_crud_api

Generate CRUD API from database schemas

cache_management

Manage cache behavior and invalidation

environment_management

Manage environment variables and configurations

manage_secrets

Manage secret values securely

manage_functions

Manage server-side functions

manage_triggers

Manage database triggers

manage_webhooks

Manage webhooks