Set it up
- 1
Turn on Neon in your AI
In Claude: open Settings → Connectors and add Neon. In ChatGPT: Settings → Apps & Connectors.
- 2
Add it to your app once
Copy the short config below into your app. It just tells your AI where to find Neon.
- 3
Just ask
Tell it what you want in plain words — no special commands.
For example, say
“List all my Neon projects and their current branches”
Need the exact config? Show it +
Add to your client
{
"mcpServers": {
"neon": {
"command": "npx",
"args": ["-y", "@neondatabase/mcp-server-neon", "start", "<your-neon-api-key>"]
}
}
}Paste into your client's MCP configuration file.
Get the skill
A ready-made skill that teaches your AI agent to use Neon well. Neon's official Postgres skills — branching, connections, migrations.
npx skills add neondatabase/agent-skills -s neon-postgres Quick skills
Copy one, paste it to your AI, watch it work.
Branch-Based Testing
“Create a branch of my production database called 'test-migration' and run the new migration there”
Multi-Branch Queries
“Compare the user table schema between the main branch and the staging branch”
Project Scaffolding
“Create a new Neon project in us-east-2 and give me the connection string”
Good to know
- · Neon API keys can be scoped to specific projects — use the narrowest scope you can
- · Branches share storage but have their own compute — spinning up too many can incur cost
- · Deleting a branch is instant and irreversible — there's no recycle bin
When to use
- ✓ You use Neon for serverless Postgres and want AI help managing it
- ✓ You need to create ephemeral branches for testing migrations
- ✓ You want to query databases across branches conversationally
- ✓ You're setting up new projects and want AI to scaffold them
When not to use
- × You self-host Postgres (use postgres-mcp instead)
- × You use Supabase, PlanetScale, or another managed Postgres
- × You need features not exposed by Neon's API (like custom extensions)
About
Imagine telling your AI: 'Create a new branch of my production database for testing' or 'Run this migration on the staging branch' — and it manages Neon for you. This server connects your AI to Neon's serverless Postgres so it can manage projects, create branches, run queries, and handle migrations through conversation.
The official Neon MCP server gives AI assistants full control over Neon serverless Postgres databases. Create and manage database branches, run queries, handle migrations, inspect schemas, and monitor usage — all through natural language.
Maintained by Neon with deep integration into Neon's branching workflow. Ideal for teams using Neon's serverless Postgres.
Workflows That Use Neon MCP
Works With
Also Consider
FAQ
What does Neon MCP do? +
Imagine telling your AI: 'Create a new branch of my production database for testing' or 'Run this migration on the staging branch' — and it manages Neon for you. This server connects your AI to Neon's serverless Postgres so it can manage projects, create branches, run queries, and handle migrations through conversation.
Do I need to know how to code? +
No. Turn it on in your AI's settings and ask in plain English — no terminal, no coding.
When should I use Neon MCP? +
Reach for it when you need to: You use Neon for serverless Postgres and want AI help managing it; You need to create ephemeral branches for testing migrations; You want to query databases across branches conversationally; You're setting up new projects and want AI to scaffold them.
When should I avoid Neon MCP? +
Skip it when: You self-host Postgres (use postgres-mcp instead); You use Supabase, PlanetScale, or another managed Postgres; You need features not exposed by Neon's API (like custom extensions).
How do I install Neon MCP? +
In Claude Code, run: claude mcp add neon -- npx -y @neondatabase/mcp-server-neon start <your-neon-api-key>. Copy-paste configs for Claude Desktop, Cursor, and Windsurf are at the top of this page.
