AI Response Truncation on Client's WhatsApp When Using Flow Builder Typebpt.io
mdenari opened this issue · comments
Description:
I have a situation where using Flow Builder that accesses an AI via API, when the AI responses include a menu of options, the responses appear correctly in the Flow Builder's conversation simulation screen, but on the client's WhatsApp, only part of the response is displayed. I understand that since the internal test chat of the Flow Builder shows the complete response, it seems that part of the AI's response is being truncated. See the attached images.
Steps to Reproduce:
- Using the Typebot Chatbot connected to the EvolutionAPI, the chatbot sends an API request to Flowise with a client's question, in the example "oi". Using the Typebot Chatbot connected to the EvolutionAPI, the chatbot sends an API request to Flowise with a client's question, in the example "hi" (highlighted in orange). The AI running Flowise with OpenAI responds in "response" and returns the text along with the choice options, as you can see in the image.i" (highlighted in orange). The AI running Flowise with OpenAI responds in "response" and returns the text along with the choice options, as you can see in the image. Notice that the entire menu is printed in the internal chat of the Typebot.
👆As you can see in the image above, within the chatbot environment, it receives the message and the response options.
👇In the client's WhatsApp connected to the EvolutionAPI, the AI response passed through the Typebot Chatbot is not fully displayed on the client-side via WhatsApp. The text is not completed.
Notice that Typebot received the entire request information, but only sent part of the text to WhatsApp.
What is highlighted in blue arrived on WhatsApp, but what is in red was not sent to the client on the frontend.
I expect the entire message received from the AI to be sent to the frontend to which the Evolution API is connected.
Environment:
Server: [e.g., Ubuntu 22.4]
API Version: is the Last Version
Other Hardware/Software Specifications: [e.g., 3 vcpus, GPU]
Resolving this issue is crucial for deciding whether to use typebot for new developments using new AI tools like Flowise, CrewAI, Dify, and Relevance.
Thank you for your kind words! We're delighted to know that we're making a difference. If you need anything else or have more questions in the future, we're here to help. Wishing you success as well!
We don't officially support Evoluation API