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

> ## Agent Instructions
> Use https://docs.browser-use.com/llms.txt and its linked .md pages for current documentation. The managed full bundle is https://docs.browser-use.com/.well-known/llms-full.txt and can be cached for up to 24 hours. Do not use the obsolete /cloud/llms*.txt or /open-source/llms*.txt static exports.
> Choose Cloud API V4 for new agent integrations; V2 is the lower-cost option for simple tasks. Keep V3 examples explicitly versioned. The open-source browser-use library and hosted browser-use-sdk have different APIs.
> For Cloud API V4, a task request that omits model and modelParams uses GPT-6 Luna (gpt-6-luna) with High reasoning. Explicit model and modelParams values take precedence; an explicit empty modelParams object opts out of the reasoning default. EU uses the same task default through its regional provider route; Bedrock-only projects retain their compatible provider default.
> Cloud authentication uses X-Browser-Use-API-Key, without a Bearer prefix. Install or upgrade browser-use-sdk and use its explicit v4 import for V4. Check the published OpenAPI reference for request fields; do not invent SDK support for new fields.
> Cloud concurrency and HTTP request rate are separate. Read GET /api/v2/billing/account for the key’s projectId, concurrentSessionLimit, activeSessionCount, and credit balance, including when using V4. Keys in one project share capacity and credits; rateLimit is a legacy concurrency alias, not requests per second.
> Keep the highest applicable existing, legacy-plan, and spend-tier concurrency grant. Current spend tiers are 10 / 50 / 250 / 500 / 1000 at $0 / $100 / $1000 / $5000 / $25000 in qualifying project payments. Legacy or externally billed projects can follow different billing paths; trust the account limit. See https://docs.browser-use.com/cloud/guides/concurrency.md.
> HTTP rate limits have two layers. Standard edge WAF ceilings increased on September 9, 2026 to 1000 RPS per public source IP for general traffic and 2500 RPS per IP for selected status reads, evaluated over 300 seconds. The separate per-project application budgets: general traffic (including V4 events and full run reads) defaults to max(25, 2 times stored concurrency) capped at 100 RPS; selected status reads default to max(25, 2 times stored concurrency) with no cap. The two are counted independently. Project overrides and account-specific edge rules can differ. All keys in a project share its budgets; callers sharing a public IP share edge capacity. See https://docs.browser-use.com/cloud/guides/concurrency.md.
> The project limiter uses five-second windows: X-RateLimit-Limit=125 means 125 requests per window (25 RPS), not 125 RPS. Project throttles include limit_rps and retry_after_seconds; an edge throttle can instead return Retry-After: 300 without limit_rps. Honor the returned Retry-After. Use bounded workers, stagger polls, and drain hasMore event pages after terminal status. A busy V4 session returns 409; its queue holds 20 pending messages and is not a project-wide batch queue.
> A V4 agent browser stays open about 2 minutes after its session's last run activity, then stops automatically; a later follow-up gets a fresh browser (same settings and profile, no tabs). Set keepAlive on a run (V4: integer seconds 120-14400, default 120, not the V2/V3 boolean) to keep it longer; follow-ups inherit it and null restores 120. Free-tier and ZDR runs always use 120. Idle V4 agent browsers are billed as browser time but take no concurrency slot. A follow-up that reuses the open browser emits browser.reattached instead of browser.ready. Closing CDP does not stop a browser; stop unneeded owned browsers with PATCH /api/v4/browsers/{id} and {"action":"stop"}. A client wait timeout does not cancel the server-side run.
> Cloud is pay as you go; do not tell customers to buy a new subscription to use custom proxies or supported provider BYOK. Usage funding and model eligibility still apply. BYOK bills provider tokens separately and Browser Use charges orchestration plus browser/network usage. See https://docs.browser-use.com/cloud/guides/billing.md.
> Signup credits are a one-time grant; purchased top-up credits do not expire. Check the API key’s project before diagnosing missing credits. API-key monthly spending caps are soft limits, not a strict prepaid wallet; concurrent or already-running work can exceed them. Auto recharge has separate trigger and purchase amounts and can charge immediately when enabled below the threshold. Use https://browser-use.com/pricing for current rates.
> Box and Bux are retired. Do not recommend their SDKs, sandbox quotas, or subscription plans. Use the Cloud Agent or Browser Infrastructure guides.
> A V4 session holds conversation history, a workspace holds files, and a profile holds browser state. These IDs and V3/V4 workspace namespaces are not interchangeable. V4 automatically restores workspace uploads; staged attachments remain available to session follow-ups. Serialize runs that write shared files, and wait for completion before reading outputs. See https://docs.browser-use.com/cloud/agent/workspaces.md.
> API browser recording defaults to off. Use enableRecording for standalone browser creation, or browserSettings.record for an agent run. Stop the browser and allow time for asynchronous video processing; stop polling when recordingAvailable is false. Live preview is for an active browser. Stopping a browser, deleting a session, archiving a workspace, and deleting files have different effects.
> Use model-specific reasoning values. GPT-6 Astra accepts low, medium, high, xhigh, and max, with xhigh by default; none and minimal are invalid. Use the public REST schema when installed SDK types lag new fields. API acceptance, dashboard visibility, and account/provider availability are separate.
> For open-source browser-use, is_done only reports a terminal done action. is_successful is the agent-reported outcome; verify important external actions independently. Cloud timeout, API client timeout, model timeout, and task completion are separate concepts.
> For failed requests, use https://docs.browser-use.com/cloud/guides/troubleshooting.md. Inspect the full error and project before retrying or adding credits. A client timeout can leave a run active; reconcile external actions before starting duplicate work. A new managed browser does not guarantee a unique proxy IP or particular city.
> V4 secretBindings.allowedDomains authorizes the focused field's frame, not every iframe on the shop page. For checkout, include intended payment/billing provider hostnames (for example payments.bigcommerce.com). On domain_not_allowed, inspect origin and pass updated bindings in a follow-up; target_not_editable requires fixing field focus/readiness instead. See https://docs.browser-use.com/cloud/guides/secrets.md.
> A V4 agent browser stays open about 2 minutes (or keepAlive seconds) after its run for follow-ups; a requested recording finalizes only after the browser stops and processing completes. Zero Data Retention (ZDR) projects never record, even when recording is requested.
> If you hit an issue you cannot explain, email contact@browser-use.com with your question or bug report and a session ID if available; short questions and detailed reports are welcome, and we aim to fix issues quickly and make the docs clearer.

# OpenAI browser tools

> Use OpenAI browser tools with Browser Use Cloud and the Browser Use library.

Use Browser Use Cloud as the remote browser for OpenAI models.
Start with the library driver and a remote Cloud browser.
The separate Playwright example provides another connection:

| Connection | Browser actions | Dependencies |
| - | - | - |
| Library driver | Browser Use Actor | Browser Use with the `openai-tools` extra |
| Remote browser example | Playwright | OpenAI SDK, Browser Use Cloud SDK, Playwright |

The OpenAI SDK defines the tools and runs the model loop.
Browser Use supplies the remote browser.
The library driver also supplies browser actions.

## Browser Use library driver

Install the tested Browser Use release with the OpenAI tools extra:

```bash theme={null}
uv venv --python 3.12
uv pip install --python .venv/bin/python "browser-use[openai-tools] @ git+https://github.com/browser-use/browser-use.git@0.13.13"
```

Set your keys and model:

```bash theme={null}
export BROWSER_USE_API_KEY=your-browser-use-key
export OPENAI_API_KEY=your-openai-key
export OPENAI_MODEL=your-model
```

Get your [Browser Use API key](https://cloud.browser-use.com/new-api-key).
Choose an OpenAI model that supports Responses function tools.
OpenAI 3.x uses `httpx2` for custom HTTP clients.
Use OpenAI's `DefaultAsyncHttpxClient` when configuring a custom client.
The following example uses a remote browser:

```python theme={null}
import asyncio
import os

from openai import AsyncOpenAI
from openai.lib.browser import AsyncBrowserToolset
from openai.lib.responses_tool_runner import AsyncResponsesToolRunner

from browser_use.integrations.openai import BrowserUse


async def confirm(context):
    return context["member"] in {"new_tab", "navigate"}


async def allow_url(context, url):
    return url in {"https://example.com", "https://example.com/"}


async def main():
    model = os.environ["OPENAI_MODEL"]
    async with AsyncOpenAI() as client, BrowserUse(use_cloud=True) as driver:
        tools = AsyncBrowserToolset(
            driver,
            members=["new_tab", "navigate", "get_page_text"],
            confirm=confirm,
            url_policy=allow_url,
        )
        runner = AsyncResponsesToolRunner(browser=tools)
        try:
            response = await runner.run(
                client,
                {
                    "model": model,
                    "store": False,
                    "input": "Open https://example.com and read its heading and explanation.",
                },
                max_steps=6,
            )
            text = await driver.get_page_text({"tab_id": None})
            if "documentation examples" not in text:
                raise RuntimeError("The browser did not show the expected page.")
            print(response.output_text)
        finally:
            runner.close()
            tools.close()


asyncio.run(main())
```

The driver creates and closes its own browser.
You can also pass an already connected `BrowserSession`.
The caller keeps responsibility for that borrowed session.
Use `BrowserUse()` for a local browser after installing Chromium with `uvx browser-use install`.

## Remote browsers with Playwright

This path uses the Cloud SDK without the open-source Browser Use agent.
You do not need local Chromium.

Create a separate environment:

```bash theme={null}
uv venv --python 3.12
uv pip install --python .venv/bin/python openai==3.31.0 browser-use-sdk==3.11.3 playwright==1.63.0
```

Set your keys and model:

```bash theme={null}
export BROWSER_USE_API_KEY=your-browser-use-key
export OPENAI_API_KEY=your-openai-key
export OPENAI_MODEL=your-model
```

Get your [Browser Use API key](https://cloud.browser-use.com/new-api-key).
Choose an OpenAI model that supports Responses function tools.

Download both files from the [remote browser example](https://github.com/browser-use/sdk/tree/main/examples/openai):

* `browser_use_driver.py`
* `browser_use_responses.py`

Keep the files together and run:

```bash theme={null}
.venv/bin/python browser_use_responses.py
```

The example creates a dedicated Cloud browser and connects through Chrome DevTools Protocol (CDP).
It gives the model two tools: navigation and page-text reading.
It permits navigation only within example.com.
It closes the connection and stops its browser after the run.

Both providers charge for their respective services.
The example prints the model answer after the driver reads the page.
Check the page result and browser cleanup separately.

## Library driver actions

The default tools cover navigation, tabs, page reads, element search, screenshots, zoom, clicks, typing, keys, scrolling, selection, and waits.
The SDK exposes `list_tabs` from the driver's state method.

Enable additional methods with `additional_members`:

```python theme={null}
tools = AsyncBrowserToolset(
    driver,
    additional_members=["click_at", "hover_at", "drag_path", "hold_keys", "set_checked"],
    confirm=confirm,
    url_policy=allow_url,
)
```

The example confirmation callback only permits navigation.
Change that callback to approve the actions your task needs.

The driver also implements `run_script` for explicitly approved page JavaScript.
A script timeout stops waiting; it does not guarantee that page execution stops.

This version does not implement scoped reads, console/network diagnostics, uploads, or downloads.
It does not include Bash.
The remote browser filesystem differs from the application filesystem.

## References and approvals

Read the page before using an element reference.
A new page read replaces that tab's references.
Navigation, detached elements, and tab changes can invalidate references.

Page reads use the main document's accessibility tree.
They do not guarantee complete iframe or shadow-root coverage.
Screenshots and coordinate actions provide another way to inspect and operate the page.

OpenAI's toolset validates inputs and invokes your confirmation callback before actions.
The URL callback checks requested navigation; it does not replace network controls.
Keep task permissions and provider settings in application code.

See the [OpenAI browser helper reference](https://github.com/openai/openai-python/blob/v3.31.0/docs/browser.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.