# Build a Neon reader with OpenAI Dots Prepare a grounded database project inventory

> Use OpenAI Dots to build a separate read-only Neon assistant with Agent200. Ground the database project inventory in returned source data.

- Page: https://www.agent200.com/connectors/neon/framework/openai-dots/
- Connector: [Neon](https://www.agent200.com/connectors/neon.md)
- Framework: OpenAI Dots
- Updated: 2026-11-17

## Instructions for AI agents

Goal: connect Neon to OpenAI Dots with Agent200.
Guide: https://www.agent200.com/connectors/neon/framework/openai-dots/

1. Get an Agent200 API key. If the developer has none, send them to https://agent200.com/platform/apikeys/ and ask them to choose "Create an API key". Read the key from the AGENT200_API_KEY environment variable. Never ask for it in chat, and never write it into code, prompts or logs.
2. Make sure Neon is enabled for the project in the Agent200 dashboard (Connectors), with the endpoints and scopes the task needs.
3. Add the Agent200 SDK to the project and create a client with that key.
4. Run requests through Agent200 with a model (for example openai/gpt-5.6-sol) and "neon" in the tools. The model decides when a Neon tool is needed; Agent200 executes the call.
5. Check it works with this request: "Read the Neon project I identify and prepare a database project inventory. Name the source and use only project, branch, database and operation fields. State access gaps; do not change records."
   What should come back:
   - The selected project is identified.
   - Only the returned project, branch, database and operation fields are reported.
   - Access gaps and missing context are explicit; no source data is changed.
6. If a tool is missing or a call is refused, check the project's Connectors settings in the dashboard, then retry.

The code on the guide page is a conceptual example of the SDK. SDK reference: https://docs.igpt.ai/

## Overview

Build a separate database project inventory application with Agent200. Use the runtime’s real schemas and preserve the authenticated user. This is an architecture guide, not a native installation in OpenAI Dots.

Start with one identified project. Preserve returned project, branch, database and operation fields; report missing data and keep the workflow read-only.

Provider reference: [Neon API](https://neon.com/docs/reference/api/projects/list-projects).

Use the provider credential described on the Neon parent page. Keep it in secure configuration and verify the selected tenant or account before reading.

- **Connector:** [Neon](https://www.agent200.com/connectors/neon/)
- **Works with:** OpenAI Dots · Your own agent code
- **Authentication:** Your Agent200 API key
- **You need:** [An Agent200 API key](https://agent200.com/platform/apikeys/)
- **Tools:** [5 tools](https://www.agent200.com/connectors/neon/framework/openai-dots/#tools)
- **Model in examples:** `openai/gpt-5.6-sol`

## Before you start

- **An Agent200 API key.** Create one on the [API keys page](https://agent200.com/platform/apikeys/) of the dashboard.
- **Neon enabled for your project.** In the dashboard's Connectors settings, with the endpoints and scopes your agent needs.
- **A model for your agent.** The examples use openai/gpt-5.6-sol, through the same Agent200 API key.

## Connect Neon to OpenAI Dots

### Step 1: Create your Agent200 API key

Create a key on the [API keys page](https://agent200.com/platform/apikeys/) and store it securely.

File `.env`:

```bash
# Keep the key out of your code, prompts and chats.
export AGENT200_API_KEY="paste-your-key-here"
```

(Conceptual example of the Agent200 SDK.)

### Step 2: Inspect the real tool definitions

Confirm the actual Neon read operations and input schemas in your runtime.

### Step 3: Configure Neon access

Use the provider-issued credential through the supported Agent200 connection route. Confirm the account scope and inspect the advertised tools before reading.

### Step 4: Build and test the read-only answer

Prepare the database project inventory only from returned source data. The code example illustrates a sequence, not an SDK installation contract.

File `run.js`:

```js
const result = await agent200
  .user("user_123")
  .run({
    model: "openai/gpt-5.6-sol",
    input: "List my Neon projects",
    tools: ["neon"],
  });

console.log(result.output);
```

(Conceptual example of the Agent200 SDK.)

## Test your connection

Ask: "Read the Neon project I identify and prepare a database project inventory. Name the source and use only project, branch, database and operation fields. State access gaps; do not change records."

What should come back:

- The selected project is identified.
- Only the returned project, branch, database and operation fields are reported.
- Access gaps and missing context are explicit; no source data is changed.

Check the selected project ID, account scope, read permission and pagination. A project inventory is control-plane metadata, not database row access. Do not expose connection strings or credentials in an agent answer.

## Neon tools OpenAI Dots can call

- **List projects** (`neon_list_projects`, read): Find projects visible to the Neon API key.
- **Read a project** (`neon_get_project`, read): Retrieve one project by ID.
- **List branches** (`neon_list_branches`, read): Find branches for an identified project.
- **Read a branch** (`neon_get_branch`, read): Retrieve one branch by ID.
- **List databases** (`neon_list_databases`, read): Find databases on the selected branch.

## How it works

1. **Your app sends the request.** It calls Agent200 with the model, instructions, the input and the Neon tools.
2. **The model picks a tool.** The model decides whether a Neon tool is needed, and with which arguments.
3. **Agent200 runs the call.** It reaches Neon with your Agent200 API key, within the endpoints your project allows, and returns the result to the model.
4. **Your app gets the result.** Agent200 returns the completed result. Your app decides when the agent runs and what happens next.

**Permissions at every level.** Your project settings decide which services and endpoints the agent may use.

## Troubleshooting

### The source cannot be read

Check the selected project ID, account scope, read permission and pagination. A project inventory is control-plane metadata, not database row access. Do not expose connection strings or credentials in an agent answer.

## FAQ

### How do I verify this result?

Compare the named project and returned project, branch, database and operation fields with the source. Treat missing information as missing, not as a fact.

### Is this a native product integration?

This describes a separate application built with Agent200 and OpenAI Dots; it does not install a connector into the OpenAI Dots app.

## Use Neon with another framework

- [Neon + Claude](https://www.agent200.com/connectors/neon/framework/claude.md)
- [Neon + ChatGPT](https://www.agent200.com/connectors/neon/framework/chatgpt.md)
- [Neon + Claude Code](https://www.agent200.com/connectors/neon/framework/claude-code.md)
- [Neon + Muse](https://www.agent200.com/connectors/neon/framework/muse.md)
- [Neon + Grok Bot](https://www.agent200.com/connectors/neon/framework/grok-bot.md)
- [Neon + Claude Duet](https://www.agent200.com/connectors/neon/framework/claude-duet.md)
- [Neon + Hermes Agent](https://www.agent200.com/connectors/neon/framework/hermes-agent.md)
- [Neon + OpenClaw](https://www.agent200.com/connectors/neon/framework/openclaw.md)

---

Agent200: https://www.agent200.com/ · API keys: https://agent200.com/platform/apikeys/ · Docs: https://docs.igpt.ai/
