Hi,
I’m working on a technical challenge related to building an AI chatbot connected to a public institution’s website. The goal is to have it respond in real time using up-to-date information (PDFs and URLs from the website), which change and update frequently.
I’m running into two key issues:
– I can’t automatically fetch and process newly uploaded PDFs
– The system doesn’t catch newly added or updated links on the site
If you have experience with Make.com, web crawling, or integrating AI with live web data, I’d really appreciate any help or direction.
Thanks a lot! 
You need to add a make.com webhook URL to the website’s API so that you can receive the realtime data
Thanks for the explanation, in the case of a Chatbase bot, my main goal is to automatically update its knowledge base regularly using Make.com (for example, when a new PDF or link is detected on the website), but I’d like to avoid any duplication or unnecessary overwriting of existing data.
The idea is to add only new links and documents without disrupting the content that’s already in the database and working correctly.
Do you have any tips on how to approach this properly through the Chatbase API? For example, how to handle duplicate checks, how the payload should be structured, or any best practices when updating the knowledge base frequently?
Thank you very much!
According to the official Chatbase documentation, there is no API endpoint to update/upload PDFs via the Chatbase API.
What you can do next:
- Please request this API endpoint to be added to the Chatbase API by the Chatbase developers first.
Hope this helps! Let me know if there are any further questions or issues.
— @samliew