How do I create a Neon API Key?

Log in to the [**Neon Console**](https://console.neon.tech), go to **Account Settings** > **API Keys** (or **Organization Settings** > **API Keys**), click **Create New**, choose the scope (Personal, Organization or Project-scoped), and copy the token immediately — it is shown only once.

What is a Neon branch and how does branching work?

A Neon branch is an isolated PostgreSQL environment with its own compute, databases and roles. Branches are created instantly using copy-on-write cloning from a parent branch (default: primary). This means zero data copy and instant provisioning. Each branch gets its own compute endpoints and can have different PostgreSQL schemas and data.

Can I get a ready-to-use connection URI?

Yes! Use the `get_connection_uri` tool with your project_id and optionally a branch_id. It returns a complete PostgreSQL connection string (postgres://user:password@host/dbname) ready to use with psql, ORMs or application clients.

Can the agent delete projects and branches?

Yes, the agent has tools to delete both projects and branches. Deleting a project marks it for recovery (7-day grace period) but destroys all associated data. Deleting a branch is irreversible and removes all its databases, roles and endpoints. Use caution when executing these actions.

Neon MCP Server for Pydantic AI 17 tools — connect in under 2 minutes

Name: Neon
Availability: InStock
Author: Vinkius

ai.pydantic.dev/mcp

Built by Vinkius GDPR 17 Tools SDK

Pydantic AI brings type-safe agent development to Python with first-class MCP support. Connect Neon through the Vinkius and every tool is automatically validated against Pydantic schemas — catch errors at build time, not in production.

Get MCP Server for AI Agents

ASK AI ABOUT THIS MCP SERVER

Open in ChatGPT Open in Claude Open in Perplexity

Vinkius supports streamable HTTP and SSE.

python

import asyncio
from pydantic_ai import Agent
from pydantic_ai.mcp import MCPServerHTTP

async def main():
    # Your Vinkius token — get it at cloud.vinkius.com
    server = MCPServerHTTP(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")

    agent = Agent(
        model="openai:gpt-4o",
        mcp_servers=[server],
        system_prompt=(
            "You are an assistant with access to Neon "
            "(17 tools)."
        ),
    )

    result = await agent.run(
        "What tools are available in Neon?"
    )
    print(result.data)

asyncio.run(main())

Fully ManagedVinkius Servers

60%Token savings

High SecurityEnterprise-grade

IAMAccess control

EU AI ActCompliant

DLPData protection

V8 IsolateSandboxed

Ed25519Audit chain

<40msKill switch

Stream every event to Splunk, Datadog, or your own webhook in real-time

* Every MCP server runs on Vinkius-managed infrastructure inside AWS - a purpose-built runtime with per-request V8 isolates, Ed25519 signed audit chains, and sub-40ms cold starts optimized for native MCP execution. See our infrastructure

About Neon MCP Server

Connect your Neon account to any AI agent and take full control of your serverless PostgreSQL infrastructure through natural conversation.

Pydantic AI validates every Neon tool response against typed schemas, catching data inconsistencies at build time. Connect 17 tools through the Vinkius and switch between OpenAI, Anthropic, or Gemini without changing your integration code — full type safety, structured output guarantees, and dependency injection for testable agents.

What you can do

Project Management — List, create, update and delete Neon projects with region and PostgreSQL version selection
Branch Operations — Create instant branches via copy-on-write cloning, set primary branches and manage branch lifecycle
Compute Endpoints — Provision read-write and read-only compute hosts for your branches
Database Administration — Create and list PostgreSQL databases within any branch
Role Management — Create database roles (users) with auto-generated passwords for secure access
Connection URIs — Get ready-to-use psql connection strings for any branch

The Neon MCP Server exposes 17 tools through the Vinkius. Connect it to Pydantic AI in under two minutes — no API keys to rotate, no infrastructure to provision, no vendor lock-in. Your configuration, your data, your control.

How to Connect Neon to Pydantic AI via MCP

Follow these steps to integrate the Neon MCP Server with Pydantic AI.

Install Pydantic AI

Run pip install pydantic-ai

Replace the token

Replace [YOUR_TOKEN_HERE] with your Vinkius token

Run the agent

Save to agent.py and run: python agent.py

Explore tools

The agent discovers 17 tools from Neon with type-safe schemas

Why Use Pydantic AI with the Neon MCP Server

Pydantic AI provides unique advantages when paired with Neon through the Model Context Protocol.

Full type safety: every MCP tool response is validated against Pydantic models, catching data inconsistencies before they reach your application

Model-agnostic architecture — switch between OpenAI, Anthropic, or Gemini without changing your Neon integration code

Structured output guarantee: Pydantic AI ensures tool results conform to defined schemas, eliminating runtime type errors

Dependency injection system cleanly separates your Neon connection logic from agent behavior for testable, maintainable code

Neon + Pydantic AI Use Cases

Practical scenarios where Pydantic AI combined with the Neon MCP Server delivers measurable value.

Type-safe data pipelines: query Neon with guaranteed response schemas, feeding validated data into downstream processing

API orchestration: chain multiple Neon tool calls with Pydantic validation at each step to ensure data integrity end-to-end

Production monitoring: build validated alert agents that query Neon and output structured, schema-compliant notifications

Testing and QA: use Pydantic AI's dependency injection to mock Neon responses and write comprehensive agent tests

Neon MCP Tools for Pydantic AI (17)

These 17 tools become available when you connect Neon to Pydantic AI via MCP:

create_branch

Optionally set a human-readable name and a parent branch ID to clone from (if omitted, clones from the project's primary branch). Branches are created instantly with zero data copy overhead. Returns the new branch along with its initial endpoints, databases and roles. Create a new branch in a Neon project

create_database

Requires the database name. Optionally set the owner role name (must exist in the branch — see list_roles). Returns the new database metadata. Create a new database in a Neon branch

create_endpoint

Specify the endpoint type: "read_write" for full access or "read_only" for read replicas. A branch can have at most one read_write endpoint. Returns the new endpoint with its connection host and configuration. Create a compute endpoint for a Neon branch

create_project

Optionally set a human-readable name, AWS region (e.g. "aws-us-east-2", "aws-eu-central-1") and PostgreSQL version (15, 16, 17). A default branch, database and read-write endpoint are automatically provisioned. Returns the new project along with its initial connection URIs, roles, databases and endpoints. Create a new Neon project

create_role

The role can be used to authenticate database connections and own databases. Provide the project_id, branch_id and desired role name. Returns the new role metadata including the generated password. Create a new database role in a Neon branch

delete_branch

The primary branch cannot be deleted — set another branch as primary first. Provide the project_id and branch_id. WARNING: this action is irreversible and destroys all branch data. Delete a Neon branch

delete_project

The project is recoverable for 7 days via the Neon console. All associated branches, databases, endpoints and data are deleted. Provide the project_id. WARNING: this action destroys all data in the project. Delete a Neon project

get_branch

Provide both the project_id and branch_id. Get details for a specific Neon branch

get_connection_uri

Optionally specify a branch_id to get the URI for a specific branch (defaults to the primary branch). The URI includes the host, database name, role and password. Use this to connect psql, ORM tools or application clients. Get a PostgreSQL connection URI for a Neon project

get_project

Provide the project_id (e.g. "purple-shape-411361") obtained from list_projects. Get details for a specific Neon project

list_branches

Each branch is an isolated PostgreSQL environment with its own compute, databases and roles. Branches can be created from any point-in-time using copy-on-write cloning. Returns branch ID, name, parent ID, primary status, creation date and current state. Use the project_id from list_projects. List branches in a Neon project

list_databases

Each database has a name, owner role and creation metadata. Use the project_id and branch_id to scope the query. List databases in a Neon branch

list_endpoints

Each endpoint has a type (read_write or read_only), host address, current state (active, idle, suspended) and autoscaling configuration. A branch can have at most one read_write endpoint. Use the project_id and branch_id. List compute endpoints for a Neon branch

list_projects

Each project is a workspace that contains branches, compute endpoints, databases and roles. Returns project ID, name, region, PostgreSQL version, creation date and resource usage metadata. Use this as the starting point for all Neon operations — you need a project_id to manage branches, databases or endpoints. List all Neon projects

list_roles

Each role has a name, creation date and privilege metadata. Use the project_id and branch_id to scope the query. Roles are used to authenticate database connections and control access. List database roles in a Neon branch

set_primary_branch

The primary branch is the default source for new branch cloning and receives the default read-write compute endpoint. Provide the project_id and the branch_id to promote. Set a branch as the primary branch of a Neon project

update_project

Provide the project_id and the new name. This does not affect branches, databases or endpoints. Update a Neon project name

Example Prompts for Neon in Pydantic AI

Ready-to-use prompts you can give your Pydantic AI agent to start working with Neon immediately.

"List all my Neon projects and show me which regions they're in."

"Create a new branch called 'feature-auth' from the primary branch of my project."

"Get the connection URI for the main branch of my project."

Troubleshooting Neon MCP Server with Pydantic AI

Common issues when connecting Neon to Pydantic AI through the Vinkius, and how to resolve them.

MCPServerHTTP not found

Update: pip install --upgrade pydantic-ai

Neon + Pydantic AI FAQ

Common questions about integrating Neon MCP Server with Pydantic AI.

How does Pydantic AI discover MCP tools?

Create an MCPServerHTTP instance with the server URL. Pydantic AI connects, discovers all tools, and generates typed Python interfaces automatically.

Does Pydantic AI validate MCP tool responses?

Yes. When you define result types as Pydantic models, every tool response is validated against the schema. Invalid data raises a clear error instead of silently corrupting your pipeline.

Can I switch LLM providers without changing MCP code?

Absolutely. Pydantic AI abstracts the model layer — your Neon MCP integration works identically with OpenAI, Anthropic, Google, or any supported provider.