---
title: "How To Install"
img: https://s3.us-east-2.amazonaws.com/unified-article-images/how_to_install-icon.png
date: 2026-01-05T00:00:00.000Z
tag: Product
description: "Agent Builder connects to Unified MCP by adding a remote MCP server URL, then enabling only the tools your agent needs."
url: "https://unified.to/blog/how_to_install"
---

# How To Install
------
_January 5, 2026_

Agent Builder connects to [Unified MCP](https://docs.unified.to/mcp/overview) by adding a remote [MCP](/mcp) server URL, then enabling only the tools your agent needs.


This is the fastest path to give OpenAI agents real-time access to customer systems through Unified connections, without building one-off function calls for each SaaS API.


## 1) Create your Unified MCP server URL


Use the Streamable HTTP endpoint:

- US: `https://mcp-api.unified.to/mcp`
- EU: `https://mcp-api-eu.unified.to/mcp`
- AU: `https://mcp-api-au.unified.to/mcp`

Required parameters

- `token`

    Your Unified workspace API key (private), or an end-user-safe token for a single connection.

- `connection`

    The end-customer connection ID (commonly used when `token` is a workspace API key).


Tokens can be passed as:

- a `token=` URL parameter, or
- `Authorization: bearer {token}` [Unified Docs](https://docs.unified.to/mcp/additional-api-endpoints)

### Scoping parameters (recommended)

- `permissions`

    A comma-delimited list of Unified scope permissions (example format shown in docs: `hris_employee_read`). [Unified Docs](https://docs.unified.to/mcp/additional-api-endpoints)

- `tools`

    A comma-delimited list of MCP tool IDs (example formats shown in docs: `get_messaging_message`, `list_messaging_message`). [Unified Docs](https://docs.unified.to/mcp/installation)

- `hide_sensitive=true`

    Removes sensitive fields from results. [Unified Docs](https://docs.unified.to/mcp/additional-api-endpoints)


Optional:

- `aliases`
- `include_external_tools=true` [Unified Docs](https://docs.unified.to/mcp/additional-api-endpoints)

### Example URL (placeholder values)


```plain text
https://mcp-api.unified.to/mcp
  ?token=WORKSPACE_API_KEY
  &connection=CONNECTION_ID
  &permissions=hris_employee_read
  &hide_sensitive=true
  &tools=get_messaging_message,list_messaging_message
```


If you haven't selected tools yet, skip `tools=` and fetch the tool list first.


## 2) Fetch tool IDs (so `tools=` is exact)


Unified exposes tool discovery and execution endpoints:

- `GET /tools` to list available tools (including the `id` field) [Unified Docs](https://docs.unified.to/mcp/additional-api-endpoints)
- `POST /tools/{id}/call` to execute a specific tool [Unified Docs](https://docs.unified.to/mcp/additional-api-endpoints)

In practice, the Tools endpoint sits under the MCP base URL. For Streamable HTTP, that means:

- `GET https://mcp-api.unified.to/mcp/tools?...` (requires auth; the endpoint returns 401 if you hit it without a token)

Workflow:

1. Call `GET /mcp/tools` using the same `token`, `connection`, and `permissions` you'll use in production.
2. Copy the `id` values you want.
3. Add them back into your MCP URL using `&tools=id1,id2,...`.

## 3) Connect Unified MCP in OpenAI Agent Builder


Agent Builder's UI evolves quickly (beta), but the flow is consistent:

1. Open your Agent Builder workflow and select your agent.
2. Go to **Tools** and add an **MCP server** (remote/web MCP server).
3. Paste your Unified `/mcp?...` URL and connect.
4. Enable only the tools you want available to this agent.

If your agent has write access (create/update/delete), pair this with an approval step in your workflow.


## Why scoping matters (tool limits, cost, security)


### Tool limits


Models perform better when the toolset is small and relevant. [Unified MCP](/mcp) can expose a very large catalog, so narrowing by `permissions` and `tools` is the default best practice. [Unified Docs](https://docs.unified.to/mcp)


### Cost (OpenAI tokens + Unified API requests)


On the OpenAI side, the MCP tool is billed via tokens used when importing tool definitions and making tool calls, with no additional per-tool-call fee from OpenAI beyond token usage. [OpenAI Platform+1](https://platform.openai.com/docs/guides/tools-connectors-mcp?utm_source=chatgpt.com)


On the Unified side, **each MCP tool call counts as 1 API request on your Unified plan**. [Unified Docs](https://docs.unified.to/mcp)


### Security


Remote MCP servers can access and return sensitive data. Reduce risk by default:

- use least-privilege `permissions`
- restrict `tools`
- enable `hide_sensitive=true` [Unified Docs](https://docs.unified.to/mcp/additional-api-endpoints)

## Appendix: what Code Export will look like (Agents SDK)


If you export your Agent Builder configuration to code, the same MCP server URL typically appears as a hosted MCP tool configuration.


```python
from agentsimport Agent, HostedMCPTool

agent = Agent(
  name="Unified-powered agent",
  tools=[
    HostedMCPTool(tool_config={
"type":"mcp",
"server_label":"unifiedMCP",
"server_url":"https://mcp-api.unified.to/mcp?token=...&connection=...&permissions=hris_employee_read&hide_sensitive=true&tools=get_messaging_message,list_messaging_message",
"require_approval":"never",
    })
  ],
)
```


The implementation stays the same across UI and SDK: connect Unified as a remote MCP server URL, then scope down to the smallest set of permissions and tools your workflow requires.


[**Start your 30-day free trial**](https://app.unified.to/login)


[**Book a demo**](https://calendly.com/d/cph9-g8n-jzg/connect-with-unified)