Skip to main content

​
Overview

A large share of the answers your customers need is already published: product documentation, getting-started tutorials, release notes, and blog posts. IrisAgent provides a specialized web scraper designed for extracting data from public websites, so that content can be ingested and used in answers without being duplicated into a knowledge base first. This is a good fit for public-facing content. For knowledge that sits behind a login or inside an internal wiki, use Confluence or Upload Content instead.

​
Add a website

  1. Go to Data Sources → Website on the IrisAgent dashboard.
  2. Open the Website Content card.
  3. Add the top-level URL to your website. This page and its subpages will be crawled and indexed.
Because the crawler follows subpages from the URL you provide, point it at the most specific root that covers the content you want. Giving it a documentation root rather than your marketing homepage keeps the index focused on pages that actually answer support questions.

​
Keeping crawled content current

The crawler re-visits your site so that newly published and updated pages are picked up over time. If you publish a page and need it reflected sooner, you can re-run the crawl from the same Website Content card.

​
Data governance and security

Data between customer environments and IrisAgent’s systems is always encrypted at rest and during transmission, authenticated at each endpoint, monitored for integrity, and protected by enterprise-grade security protocols.
  • IrisAgent only crawls pages that are publicly reachable on the URL you provide.
  • IrisAgent removes PII data and does not store any PII data on our servers.
  • IrisAgent does not share any data with third parties.
  • IrisAgent is SOC 2 Type II compliant and GDPR compliant.