Create each key on its own page, paste it below, then copy the finished lines into your project's .env file. What you paste stays in this tab. It is never sent to Wirestack or saved.
Revision 1, checked Nov 16, 2026. Installs exactly browser-use 1.0.5, creates the files and adds the key names to your .env. It shows everything first and asks before changing anything.
Adapt to my stackProShowing the original (Python). Pick your stack to see it rewritten.
Free accounts can open one adapted integration a month. Sign in to try it.
A Python script that hands a plain-English task to a Browser Use agent. The agent opens a real browser, clicks and reads its way through the web, and prints the answer. With only a Browser Use key, it runs on their hosted model (bu-fast) and a cloud browser that stops when the run ends.
Put your key in the environment (a .env file in the same folder with BROWSER_USE_API_KEY=bu_... also works):
Shell
export BROWSER_USE_API_KEY=bu_...
Save this as agent.py:
Python
from browser_use import Agentresult = Agent('Find the number of stars of the browser-use repo').run_sync()print(result.final_result())
Test it
Run python agent.py. After the agent finishes browsing, it prints the star count of the browser-use GitHub repo. Change the text inside Agent('...') to give it any other task. Each run reports its model cost in result.cost_usd.
Not confirmed
The official quickstart pins browser-use==1.0.0. This integration pins 1.0.5, the newest version on PyPI today. GitHub already shows a 1.0.6 release that is not on PyPI yet.
We did not run the script. How long a run takes and what it costs depend on the task.
To use your own model and a local Chrome instead of the cloud, the docs show Agent(..., llm=ChatOpenAI('gpt-6-luna'), browser=Browser('chromium')), which reads OPENAI_API_KEY. We did not test that path.