> For the complete documentation index, see [llms.txt](https://docs.hex-rays.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.hex-rays.com/editions-and-licenses/license-types/ida-pro-headless/concepts/headless-seats.md).

# Headless seats

IDA Pro Headless is licensed with **headless seats**: seats on your license server that are taken by scripts and agents rather than by people. This page explains when a seat is taken, when it is given back, and how many you need.

## Human seats and headless seats

Your license server can hold two kinds of licenses side by side:

* **Human seats**, for analysts working in IDA Pro. With a [floating license](https://docs.hex-rays.com/editions-and-licenses/license-types/floating-license/overview.md), an analyst takes a seat when they work in IDA, and the seat goes back to the pool when they close it.
* **Headless seats**, for anything that is not human: scripts, pipelines, AI agents, services. They work exactly the same way. A script or agent takes a seat while it runs, and gives it back when it stops.

Both kinds are served by the same license server, so your analysts and your agents work together in one environment.

## When is a seat taken?

A headless seat is taken when a script or agent **opens a database**, and given back when it **closes the database** or the process ends. A script that is not running holds no seat.

* With idalib, `open_database()` takes the seat that `IDA_LICENSE` names, and `close_database()` gives it back. A script can also take the seat before it opens a database: see [Wait for a free seat](https://docs.hex-rays.com/editions-and-licenses/license-types/ida-pro-headless/how-tos/wait-for-a-free-seat.md#or-take-the-seat-first).
* With `idat`, the seat is held from start to exit.
* With [IDA MCP](https://docs.hex-rays.com/add-ons/mcp/overview.md), each open database holds a seat until the agent closes it.

{% hint style="info" %}
Always close the database when you are done: a database left open keeps its seat busy for everyone else.
{% endhint %}

## How many seats do you need?

You need as many seats as databases open **at the same time**, not as many as scripts or binaries.

| What runs                             | Seats needed |
| ------------------------------------- | ------------ |
| 10 scripts, one after the other       | 1            |
| 10 scripts at the same time           | 10           |
| 4 agents, each with one database open | 4            |
| 1 agent with 3 databases open at once | 3            |

When every seat is in use, the next script cannot open a database until a seat is free. See [Wait for a free seat](https://docs.hex-rays.com/editions-and-licenses/license-types/ida-pro-headless/how-tos/wait-for-a-free-seat.md) to let scripts queue rather than fail.

{% hint style="success" %}
Start small and scale bottom up: take a few headless seats, watch how many your scripts and agents use at peak in the license server console, and grow the pool from there.
{% endhint %}

## Machines are not tied to a seat

Headless seats are not bound to a machine: the license server does not check the MAC address or the identity of the host. It counts how many seats are in use at the same time. Any of your machines that can reach the license server can take a free seat.

## Naming seats

Add `instance=<name>` to `IDA_LICENSE` to give each script or agent a name:

```bash
export IDA_LICENSE=server:host=licsrv.example.com:port=65434:instance=triage-worker-1
```

The name shows next to the seat on the **Allocations** page of the license server administration console, so you can see which workload holds which seat.

## Offline machines

A machine that cannot reach the license server uses a seat **borrowed** for it. A borrowed seat counts as in use for its whole borrow period, until it is returned or the period ends. See [Work offline with a borrowed seat](https://docs.hex-rays.com/editions-and-licenses/license-types/ida-pro-headless/how-tos/work-offline.md).


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.hex-rays.com/editions-and-licenses/license-types/ida-pro-headless/concepts/headless-seats.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
