> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pylar.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenAI Platform

> Connect Pylar MCP tools to OpenAI's platform for AI agent development

## Overview

OpenAI Platform provides tools for building AI agents and assistants. Connect your Pylar MCP tools to give your OpenAI agents access to your data through governed views.

## Prerequisites

* ✅ OpenAI Platform account
* ✅ Access to OpenAI agent builder features
* ✅ Pylar project with published MCP tools
* ✅ MCP HTTP Stream URL and Bearer Token from Pylar

## Step 1: Get Your Pylar Credentials

1. In Pylar, navigate to your project
2. Click **"Publish"** in the right sidebar
3. Generate your token if needed
4. Copy both:
   * **MCP HTTP Stream URL**: `https://mcp.publish.pylar.ai/mcp`
   * **Authorization Bearer Token**: Your unique token

## Step 2: Connect Pylar to OpenAI Platform

1. In the OpenAI Platform, navigate to your agent builder
2. Look for **"Connect"** or **"Add Tool"** button
3. Click **"Connect"** to add your MCP server

## Step 3: Configure MCP Connection

When you click "Connect", you'll need to provide:

* **MCP Server URL**: `https://mcp.publish.pylar.ai/mcp`
* **Authorization Token**: Your Bearer Token from Pylar

After connecting, you'll see your MCP tools appear in the platform.

## Step 4: Configure Tool Access

Once your MCP tools appear, you can:

1. **Modify Approval Status**: Set which tools require approval before use
2. **Select Tools**: Choose which tools you want the agent to have access to
3. **Configure Permissions**: Set access levels for different tools

Click the **"Add"** button to confirm your selections.

<Tip>
  You can selectively enable tools based on what your agent needs. Not all tools need to be available to every agent.
</Tip>

## Step 5: Test Your Tools

1. Click the **"Preview"** button in the top right corner
2. Ask a question that requires the agent to use your MCP tool

**Example**: *"Get engagement scores for login events"*

The agent will:

1. Recognize it needs to use your Pylar tool
2. Call the tool with appropriate parameters
3. Execute the query against your view
4. Return the results and use them to answer your question

<Check>
  The Preview feature lets you test how the agent uses your MCP tools before deploying to production.
</Check>

## Using Your Tools

Once connected and configured, your OpenAI agents can:

* Access your data through governed views
* Use tools automatically when needed
* Combine multiple tools for complex queries
* Provide answers based on real-time data

## Tool Management

### Approval Status

You can set approval requirements for tools:

* **Auto-approve**: Tools run automatically
* **Require approval**: Tools need manual approval before execution

### Tool Selection

Select which tools are available to your agent:

* Enable only tools your agent needs
* Disable tools that aren't relevant
* Update selections anytime

## Benefits

* ✅ **OpenAI Integration**: Use OpenAI's powerful models with your data
* ✅ **Selective Access**: Choose which tools agents can use
* ✅ **Approval Controls**: Require approval for sensitive operations
* ✅ **Centralized Updates**: Update views in Pylar, agents get updates
* ✅ **Full Visibility**: Monitor usage through Pylar Evals

## Troubleshooting

### Issue: Tools not appearing after connect

**Solutions**:

* Verify Bearer Token is correct
* Check MCP URL is correct
* Ensure tools are published in Pylar
* Refresh the connection

### Issue: Agent can't use tools

**Solutions**:

* Verify tools are selected/enabled
* Check approval status settings
* Review tool permissions
* Test tools in Preview mode first

## Next Steps

* [Publishing Tools](/learn/publishing-tools/publishing-your-tools) - Publish your tools
* [Evals](/learn/evals/overview) - Monitor agent usage
* [Other Builders](/learn/connecting-agent-builders/overview) - Connect to other platforms

<Card title="Monitor Usage" icon="chart-bar" href="/learn/evals/overview">
  See how OpenAI agents use your tools through Evals
</Card>