Catalog Web Import Agent

poetry add enthusiast-agent-catalog-web-import

The Catalog Web Import agent accepts one or more product page URLs and extracts structured product data directly from those pages. It uses browser-level TLS impersonation to fetch pages reliably, passes the content through an LLM to pull out the fields defined in your schema, and upserts all products in a single batch. When confirmation mode is enabled, it presents the extracted data for review before writing anything to the catalog.

Use Cases

• Supplier Website Onboarding – When a supplier shares their product catalog as a website rather than a document, the agent can scrape and import all relevant product pages directly, eliminating manual copy-paste.
• Catalog Expansion from Distributor Pages – Sales teams can drop in a list of distributor or manufacturer product URLs and have all products extracted and added to the catalog in one go, without leaving the chat interface.
• Bulk Multi-URL Import – Provide dozens of product page URLs in a single request and the agent fetches them in parallel, reconciles any overlapping data, and upserts everything in one batch call.

Installing Catalog Web Import Agent

Run the following command inside your application directory.
If you're using Enthusiast Starter, that's inside enthusiast-starter/src/

poetry add enthusiast-agent-catalog-web-import

Then, register the integration in your config/settings_override.py.

AVAILABLE_AGENTS = [
    'enthusiast_agent_catalog_web_import.CatalogWebImportAgent',
]