Websites
Enjo AI Agents can train on data from any website.
Step 1: Login to the Enjo dashboard - click on the AI Agent Studio menu item. Click on the AI agent, head over to the 'Knowledge' tab, click on the ‘Add data’ option, and choose 'From website'.
Step 2: You can select from an existing knowledge source, or create a new one.
Step 3: When creating a new website knowledge source with the following options:
- Extract Data from All Website Pages: This option allows you to retrieve data from your entire website, including internal cross-links, with a limit of 10,000 pages.
- Extract Data from Selected Pages: This enables knowledge extraction from specified web pages. Click 'Next' and select the pages you wish to index.
- Extract Data from Sitemap URL: Data will be extracted exclusively from the provided sitemap link.
Step 4: Enter a name for your knowledge source and fill out the metadata (key-value pairs) that describes this knowledge source. Metadata enhances the quality of search results, but if uncertain, you can leave it blank. Enjo will now learn from the website contents. You will be able to see the indexed documents under the 'Knowledge Sources' tab.
- What is the page limit for data extraction? The web crawling limit is set to 10,000 pages for extracting data across your website.
- Is it possible to extract data from multiple sitemaps? Currently, extraction is limited to one sitemap URL at a time.
- Will the AI Agent automatically update with the latest website data? You will need to perform periodic updates manually to ensure the agent is accessing the latest data.
- Are private web pages also crawled? The AI Agent only crawls and indexes pages available to the public unless appropriate authentication is set up.
