Indexes the coding-agent sessions already sitting on your disk — months of history from before you installed it — and serves them back over MCP. 3.5 GB searched in ~1.5 ms, no LLM, no embeddings.
Neon
Summary
Neon's official MCP server — manage serverless Postgres projects, branch databases, and run migrations.
Features
- Create and manage serverless Postgres projects and branches
- Run SQL and apply migrations safely on a branch first
- Database branching for testing schema changes
Installation
Set up this MCP server in your favorite AI agent — copy a ready-made configuration below.
Description
Neon's MCP server exposes its serverless Postgres platform to an AI assistant, and the branching model makes it unusually well suited to agent workflows.
What it enables
- Project and branch management. Create databases and branches on demand, which is fast and cheap on Neon because branches are copy-on-write.
- Safe migrations. The standout workflow: an assistant applies a schema change to a branch, verifies it works, and only then promotes it to the main branch. Getting a migration wrong costs a discarded branch rather than a production incident.
- SQL execution against any branch, with schema inspection
- Connection string retrieval for wiring the database into an application
Why branching matters here
The main risk of letting an agent touch a database is that mistakes are permanent. Branching converts that into a reversible operation, which changes what you are comfortable delegating.
Runs with npx @neondatabase/mcp-server-neon start <api-key>, or as a remote server for clients that support it.
Related MCP Servers
Weaviate's MCP server, built into the database itself — hybrid search, schema inspection, tenant listing and batch upserts at /v1/mcp, scoped by RBAC.
Google's open-source MCP server for enterprise databases — one binary that fronts AlloyDB, BigQuery, Spanner, Postgres, MySQL, Oracle, MongoDB, Snowflake and more.
A universal database MCP server — one integration for PostgreSQL, MySQL, SQL Server, SQLite, and MariaDB.
