Skip to content

Open-source Python library for AI browser agents, plus Browser Use Cloud: stealth browsers and a hosted agent API.

Example

Magnus Müller9b7bb239857e
diff --git a/examples/code_use_extract_products.py b/examples/code_use_extract_products.py
index 84af1fb22..8b9c00926 100644
--- a/examples/code_use_extract_products.py
+++ b/examples/code_use_extract_products.py
@@ -26,24 +26,13 @@ Laminar.initialize()
async def main():
- """
- Task: Extract product listings from an e-commerce site.
-
- The agent will:
- 1. Navigate to the products page
- 2. Extract products using JavaScript
- 3. Scroll to load more products
- 4. Combine all results
- 5. Save to a JSON file
- """
+
task = """
Go to https://www.flipkart.com. Continue collecting products from Flipkart in the following categories. I need approximately 50 products from:\n\n1. Books & Media (books, stationery) - 15 products\n2. Sports & Fitness (equipment, clothing, accessories) - 15 products \n3. Beauty & Personal Care (cosmetics, skincare, grooming) - 10 products\nAnd 2 other categories you find interesting.\nNavigate to these categories and collect products with:\n- Product URL (working link)\n- Product name/description\n- Actual price (MRP)\n- Deal price (current selling price) \n- Discount percentage\n\nFocus on products with good discounts and clear pricing. Target around 40 products total from these three categories.
"""
- task = "Usec requests to get GitHub's API for trending Python repos, then visit the top 3 repos in the browser and extract their README"
-
# Create code-use agent (uses ChatBrowserUse automatically)
agent = CodeAgent(
task=task,
@@ -80,11 +69,6 @@ Go to https://www.flipkart.com. Continue collecting products from Flipkart in th
notebook_path = export_to_ipynb(session, 'product_extraction.ipynb')
print(f'\n✓ Exported session to Jupyter notebook: {notebook_path}')
- # Export to Python script
- script = session_to_python_script(session)
- await asyncio.to_thread(Path('product_extraction.py').write_text, script)
- print('✓ Exported session to Python script: product_extraction.py')
-
finally:
await agent.close()