Building ChatGPT AI apps visually with Flowise: website FAQ scraper and embed chatbot in website 

on the conversational retrievals QA block, there is MEMORY node! on you example it didn't appear. to which memory block it should be connected in this example? thanks

What’s the difference between using Flowise and using StackAI? StackAI costs over $100 per month, is there anything it can do thay Flowise can’t? Thanks!

I am very ignorant in the field of AI still. Does using OpenAI as the chat model open your data up to the web? That request is going somewhere I imagine. Don't they store/log it for a certain amount of time?

Any idea on how can we prevent hallucinations in this chat bot and make it such that it answers doesn't go outside the document that we provided. Please help with any suggestions as I'm unable to find a solution for this.

Great video! Love it very cool - was looking for this quite a long time. Just one question about this Tuto: Both the Pinecone Load Existing Index and Upsert Document are DEPRACATING. So I used the NEW Pinecone available to upsert. What do I dor when I need to replace it just to load the index (instead of Upserting angain?). Thanks so much for your clarity in this tuto!!! Very usefull for newbies like me...

superb!! exactly what I was lookingfor for 2 weeks !! thank you so much for sharing and making it so easy even for a 60 yo noob!!..really appreciate it...

I have an error. Ending node must be either a Chain or Agent

Way better and easier and cheaper than using the azure open ai environment

Great video, thanks for this ! Can the chatbot also share the source inside a button or text (with previous) ? Thanks !

Sir how to hide the flowise branding powered by flowwise

Derek, How did you manage to get flowise respond so fast ? I have deployed flowise on render, and it takes around 5 seconds to get a reply from OpenAI. Other tutorials when showing how to work with flowise also have this delay. thanks

4:20 ?? Namespace? What should I write here?

Hey Derek, thanks for the tutorial. I tried following your exact steps and continue to get this error: TypeError: Cannot read properties of undefined (reading 'replace'). Any ideas why this might be happening? I tried all 3 scrapers as well as Apify. Per Apify it is successfully crawling the page but pinecone isn't receiving the data. Pinecone is receiving data in other workflows it just seems to not be receiving it from the scrapers. (Browser- Google Chrome). Thanks!

Problem with returned source buttons is that they are not Userfriendly. What I expect is just links to original documentation, but not popup with json data

Cool, but I find that many web scrapers can’t read web sites with javascript or other dynamic code. I wonder if there are web scrapers that can wait until a specific xpath shows up and then start scraping. I can do that in python with something like chrome drivers modules.

How can we scrape the dropdown filters within the same url?

Is it possible to give flowise a rotating proxy ip adress?

Thanks for this video! I'm unfortunately seeing the "TypeError: Cannot read properties of undefined (reading 'replace')" error whenever trying to to the chat. I've tried many different links in the Cheerio Web Scraper + researched elsewhere but not sure what the issue is. Would you have an idea of what I might be doing wrong?

would this work with websites like Linkedin or twitter, to search a specific subject?

does it work with a free pinecone account? Thanks

sir i have error on failed getting project named

fantastic! Love this - is there a way to temporarily disable chatbots in flowise? I can add a password to stop random usage but was hoping to either temp disable the bot as needed and/or add an open api quota to it - THANKS

Hello, The great and neat video! I am concerned about exposing such details like chatflowid: when you embed them into HTML file after deployment. People can see this and copy-paste it to use somewhere else. Is there a way to hide it? Or Do you think it is okay to be exposed?

is cheerio better than puppeteer and apify?

thanks for making this

THANK YOU FOR MAKING THIS VIDEO

How to find your pinecone namespace?

They even have FAQ. 🤯 I’m just really stuck on whether those questions are actually frequent? 😩 also do they have a menu on the website for navigating the site’s menu?

How did you deploy flowise to server online?

720p? Really?!?!

Superb! Website crawl ll happen once ? If website content is changed then how this bot aware of that and update the vecotor db?

Very useful bro👏

Mate thank you heaps for the awesome video, also please see if you able to do a video setting this up LocalAI that would be gr8.. also just wondering if anyone had any issues with these errors, Error: Request failed with status code 401,TimeoutError: The operation was aborted due to timeout. also will this work with gpt 3.5 api key can someone point me towards the right direction?

Great video. Thanks. Let's say that I want to use my own chat component that supports other features. How Can I "connect" my chat component to the Flowise flow?

Would this be scalable commercially?

hello brother when I put the condition of workflow.makeSuccess=== true it tells me that this card is ignored because it is not connected to another node even though I already entered the data in my first row of the spreadsheet and the zapier, but above when executing the code I put only workflow.makeSuccess and in const I put the zapier link but it still does not recognize makeSuccess

You can't remove flowise branding 😢