Skip to main content

Introduction

CrawlChat is a platform where you can connect your documentation from different sources and get a AI chatbot that is trained on your documentation that you can connect to your website, or as a Discord, Slack bots where people can use it to get answers for there queries, or even as an MCP server so that your community members can fully integrate your documentation into their AI apps such as Cursor, Claude Code, and more.

Process​

The process of CrawlChat workflow consists in 4 major steps at high level.

  1. Add your documentation
  2. Customise the bot
  3. Connect the channels
  4. Analyse

Let us briefly understand about those three steps now.

1. Add your documentation​

First step to start with is to add your existing documentation to the CrawlChat collection you would create after you sign up. CrawlChat covers all the major source types to building your Knowledge base that will be used to answer the queries in the later stages. Here are few

  • Scraping websites
  • Upload files
  • Github Issues
  • Notion pages

2. Customise the bot​

Once you have added your documentation, you need to customise the bot so that it works as you expected. This can be done with multiple options like

  • System prompt
  • Branding
  • AI models
  • Sample questions
  • Minimum scores

3. Connect the channels​

You can connect to different channels that covers requirements for most of the tech companies. Few of them are

  • Embed to web
  • Discord bot
  • Slack app
  • MCP server

4. Analyse​

The best part about CrawlChat is the analysis it provides from the queries it answers, the type of questsion people ask across the channels. More about it here.