SchemaFlow MCP Server

What is SchemaFlow MCP Server about?

SchemaFlow MCP Server delivers live schema metadata from PostgreSQL and Supabase databases to AI‑enhanced development environments. By exposing tables, columns, relationships, functions, triggers, enums, and indexes through MCP, AI assistants can generate context‑aware code, queries, and suggestions without needing direct database access.

How to use SchemaFlow MCP Server?

1. Create an MCP token – Sign in to the SchemaFlow dashboard, connect your database, and generate a token.
2. Add the server to your AI‑IDE – Configure the MCP endpoint (e.g., https://api.schemaflow.dev/mcp/?token=YOUR_TOKEN) in Cursor, Windsurf, or VS Code + Cline using the provided JSON snippets.
3. Invoke MCP tools – Within the IDE, ask the AI assistant to run commands such as get_schema, analyze_database, or check_schema_alignment.

Key features of SchemaFlow MCP Server

• get_schema – Retrieves full schema details or filtered subsets (tables, columns, relationships, etc.).
• analyze_database – Provides performance insights, security assessments, and structural recommendations.
• check_schema_alignment – Validates schema against best‑practice guidelines and reports actionable issues.
• Secure token‑based access – Only cached metadata is exposed; no actual data is transferred.
• Encrypted SSE communication – Real‑time streaming of schema information.
• Multi‑IDE support – Ready‑to‑use configuration for Cursor, Windsurf, and VS Code + Cline.

Use cases of SchemaFlow MCP Server

• AI‑assisted code completion – The IDE knows exact table/column names, producing correct SQL snippets.
• Automated database documentation – Generate up‑to‑date Markdown, JSON, or Mermaid diagrams on demand.
• Security & performance reviews – Run analyze_database to surface risky configurations or bottlenecks.
• Schema validation – Ensure naming conventions and relational integrity before deploying changes.
• Rapid onboarding – New developers can query the live schema via their AI assistants without digging through raw SQL.

FAQ from SchemaFlow MCP Server

Q: Does the server expose actual data? A: No. It only streams schema metadata; data rows are never transmitted.

Q: What databases are supported? A: PostgreSQL and Supabase (which is built on PostgreSQL).

Q: How is authentication handled? A: Each user receives a unique, revocable token that limits access to their own cached schema.

Q: Do I need to run any server locally? A: The MCP endpoint is hosted by SchemaFlow (https://api.schemaflow.dev). You only need to configure the URL and token in your IDE.

Q: Can I customize the information returned by get_schema? A: Yes. Provide the optional query_type parameter to filter for tables, columns, relationships, functions, triggers, enums, indexes, or all.

Q: What if I want to self‑host? A: The repository contains the server code; you can deploy it yourself and point your IDE to your own instance, using the same token‑based authentication pattern.

SchemaFlow MCP Server

Real-time PostgreSQL & Supabase schema access for AI-IDEs via Model Context Protocol (MCP)

SchemaFlow MCP Server provides your AI-IDE with live access to PostgreSQL and Supabase database schemas through the Model Context Protocol. Get real-time schema context for smarter code generation in tools like Cursor, Windsurf, and VS Code + Cline.

🚀 Quick Start

Visit SchemaFlow →

1. Get Your MCP Token

Visit SchemaFlow Dashboard to:

• Connect your PostgreSQL or Supabase database
• Load and cache your schema
• Generate an MCP token

2. Configure Your AI-IDE

For Cursor IDE:

1. Open Cursor Settings
2. Go to Features → MCP
3. Click "Add New MCP Server"
4. Use this configuration:

{
  "name": "schemaflow",
  "type": "sse",
  "url": "https://api.schemaflow.dev/mcp/?token=your-token-here"
}

For Windsurf IDE:

Add to your MCP configuration:

{
  "mcpServers": {
    "schemaflow": {
      "type": "sse",
      "url": "https://api.schemaflow.dev/mcp/?token=your-token-here"
    }
  }
}

For VS Code + Cline:

Configure in Cline's MCP settings:

{
  "mcpServers": {
    "schemaflow": {
      "type": "sse", 
      "url": "https://api.schemaflow.dev/mcp/?token=your-token-here"
    }
  }
}

🛠️ Available MCP Tools

get_schema

Retrieves complete database schema information including tables, columns, relationships, functions, triggers, enums, and indexes from PostgreSQL or Supabase databases.

Parameters:

• query_type (optional): Filter specific information (tables, columns, relationships, functions, triggers, enums, indexes, all)

Example AI queries:

• "Show me my database schema"
• "What tables do I have?"
• "Show me all relationships"
• "List database functions"

analyze_database

Performs comprehensive database analysis including performance insights, security assessment, and structural recommendations for PostgreSQL and Supabase databases.

Example AI queries:

• "Analyze my database performance"
• "Check database security"
• "Review database structure"
• "Give me a database overview"

check_schema_alignment

Validates your PostgreSQL or Supabase schema against best practices and identifies potential issues with actionable recommendations.

Example AI queries:

• "Check schema alignment"
• "Validate my database"
• "Any schema issues?"
• "Check naming conventions"

🔧 How It Works

1. Schema Caching: When you load your schema in SchemaFlow dashboard, it's automatically cached for MCP access
2. Secure Connection: Your AI-IDE connects using MCP with secure token authentication
3. Real-time Access: Your AI assistant can query schema data in real-time for better code generation

🔒 Security

• Schema metadata only - No actual data is accessed or stored
• Token-based authentication - Each user has a unique, revokable token
• Encrypted connections - All MCP communication is encrypted
• User-specific access - Tokens only access your own cached schema data

📚 Complete Documentation

For detailed setup instructions, troubleshooting, and advanced configuration, visit:

Complete MCP Integration Guide

🌐 SchemaFlow Platform

This MCP server is part of the larger SchemaFlow platform which also provides:

• Interactive Schema Visualization - Explore your PostgreSQL/Supabase database structure through intuitive diagrams
• Multi-Format Export - Export schemas in JSON, Markdown, SQL, and Mermaid formats
• Schema Browser - Navigate through tables, relationships, and database components
• Performance Analysis - Get insights into your database structure and optimization opportunities
• Supabase Integration - Native support for Supabase projects with direct connection

Visit SchemaFlow →

🆘 Support

• Documentation: schemaflow.dev/mcp-guide
• Dashboard: schemaflow.dev/dashboard
• Issues: Create an issue in this repository

📄 License

MIT License - see LICENSE file for details.