hello - I deployed a new Lindy Sales bot to my website so that it can answer questions, qualify prospects and book meetings. It's really quite snazzy in my opinion.
I noticed when I am engaging the new chatbot on my website, it says "Test Mode". I can't see to get rid of the notice. What am I missing or this a bug?
Also, the ellipsis popup is not displaying correctly.
Here are the screenshots of the Lindy Chatbot.
Hey Scrott! Thanks for reporting! The test mode is only displayed when you’re logged into Lindy. Can you share an example task url + a link to the website where its embedded? Happy to look into this. 🙏🏾 How to share your Lindy task URL: 1. Click your Lindy agent → click “tasks” 2. Find the task you need help with 3. Click the specific task (you should see the full execution) 4. Copy that page URL from your browser 5. Paste the URL in your support request Note: The URL should include “conversationID” followed by a number 💡 Screenshots are also helpful!
Marvin A. - here you go. It is doing this for all chatbot conversations. https://chat.lindy.ai/marketplace/?templateId=67e1d9734253cc39b5ebdfc1 Does this work? I added screenshots above in the replies.
hmm that does not help me to look into it. 😞 Seeing it embedded on your website — I could figure out if this was a compatibility issue or an actual bug since I can’t reproduce what you mentioned on my end.
Btw —looking at the current template you shared some potential issues that caught my eye:
the agent has too many skills. if this results in reliability issues I’d recommend lowering the number of skills to 3-4 and use some conditional logic instead
You set it to “Let AI decide” regarding confirmation. Could work but again… primer for reliability issues down the road 😄 LLM’s are pretty smart today but don’t always make best decisions
Interesting - I used Sam Wilcox's youtube video here as inspiration for my website SDR chatbot.
Followed it super closely as well
here is the website - https://optimumdata.solutions/
you are welcome to try it out. The Ellipsis issue is still happening. The Test Mode in incognito mode is now gone, which is great
yeah — not saying it won’t work. Just saying it might come with reliability issues down the road 😄 thx for sharing
what do you recommend for an alternative to "Let AI decide"?
I’d separate critical steps like deleting/updating events in my calendar — and put it on some conditional logic branch + if you let AI decide wether it needs confirmation or not it might end up waiting for your confirmation mid-conversation with a customer…. so you probably want to make sure it can end conversations too.. here is a starter template to give you an idea. Also a quick guide on our agents
one more thing - how about this bug?
Yes, one of our engineers worked on a fix and realized this will actually take some time. → team is working on it. In the meantime— the reason why this happens is because of the length of the conversation starters. You could shorten them as a quick fix for now. (Not ideal, I know 🫠 )
thanks for the update. I'll try that workaround. Thanks!