Hi! Have chatbot a memory or context understanding? Maybe I need turn on this feature…
I am asking because chatbot is stupid. For example, my first question was “What is best beaches on Phuket”. My second question was “advice me beaches for kids”. Certainly I asked about Phuket. But bot response was about Shardzha (Egipt).
Model is 4o-mini.
]]>How do you setup the Sentential Context Model –?BETA FEATURE OFFERING
]]>Hi!
First of all thank you so much for this great plugin for embedding Open Ai assistants!
I have a question about Image generation. Each time I ask it to generate an image from the embedded assistant, it takes time as if it was processing, but then nothing comes out, just a small little shape as you can see here:
https://imgur.com/a/H3Ehq3M
Even though I checked Dall-e in the backen, and enabled it in the assistant. By the way when I try the assistant in the open ai playground, it generates everything correctly. So I don’t know if i’m doing something wrong maybe?
Any help would be appreciated. TYSM
]]>I use the shortcode on the elementor:
[chatbot-1 style=embedded]
and my chat it’s not inside the elementor row but it’s going to the right corner website.
Before update to the latest version plugin the chat was on the page with full width
I use 8.1 php version on the server and all plugins and wordpress have latest version.
]]>I had the chatbot on our website but it disappeared. And now neither of the short codes work.
I used: [chatbot], [chatbot style=”floating” model=”gpt-4″].
I had the chatbot on a widget and then on our WP Bakery page builder editor. And now I don’t know what happened. I tried to contact the Bakery page builder but they aren’t available to be contacted without showing proof of license and I do not have that information since I am only an employee. If someone could help me with this. I’m sure I am missing something.
The chat bot was installed and working perfectly for about a week. No plugins were added nothing was updated but one day the bot started repeating itself using the first line of my prompt and will not move on. The line is “thank goodness, my calendar just pinged me but I didn’t want to disturb you. Are you still in the market for [niche product]”
This is the first reply then the bot is supposed to move on and qualify the lead with several more questions. The problem is also universal because every assistant I’ve set up uses a similar prompt with that first line. The only difference is the qualifying questions. However, on all my bots it’s stuck on that opening line.
Ps. also tried in incognito window to check if it was a cache issue. same issue on all bots. It says the first part line of the prompt right then repeats itself no mater what I type. I have been able to trick the bot into giving me the response that’s supposed to be triggered if a lead is mad/upset when I curse at it so that makes things even more confusing because It seems to still be referencing my Prompt somehow.
Pss. I played with things and fixed the issue on my PC but 3 others on my team are still experiencing the issue. one is in Vancouver BC, another is in Dallas, TX, another in Spain. They all still have the issue but I don’t. A mystery indeed! My team can’t sell without a working kognetics plugin. Please help!
Hey I Have a Problem. Just installed that plugin according to a tutorial. In the messages tab I get a success notification. Got 5$ on the OpenAI account.
I added the shortcake chatbot-1 on my page and it gets displayed but if I enter a message I cannot klick send
would be great if anyone knows how to fix that
]]>Hi, I’m a new gpt assistant user. I was happy to get my assistant to float on my page with Kognetiks.
However, once the answer is provided, it doesn’t open a ‘next response’ option and if the answer is longer, it doesn’t allow you to scroll. As you can see in the screen shot, the answer is long, but the scroll doesn’t go further. I have my tokens set at 2000. But no matter how high I have taken them, this still happens.
I’m sure its in my settings, but I can’t figure it out.
]]>Hi all,
the plugin seems very capable, but I am experiencing 2 problems:
Any help would be very appreciated. Thanks in advance!
]]>Hi, the link is only available to users in WP. If you want, contact me and I will give you access.<audio autoplay=””></audio>
So, for some reason chatbot stopped producing anwers according to instructions and started to create audio, and also adding the following ul list with targeting my homepage. I tried to create a GPT assistant to use that instead of primary one as I thought something is wrong with it, but the results are the same. and I noticed that my updates on assistant are not being applied on the page. Meaning: If I change the initial message and save it, it does not show up, the previous remains. I am not sure what is going on. I really dont want to use another plugin as this one is alread configured and worked well untill now.
Please help me out.
]]>This plugin is the closest to what I was looking for and I was really excited, but I’m not able to use Assistant.
I did it according to the manual, I inserted the shortcode [chatbot style=”embedded” assistant=”asst_… and the name of my Assistant appears correctly, but the conversation is as if I were using a normal chatgpt. It does not invoke the files or urls of the retrieval of my assistant.
Another thing I would like to know is if you intend to insert a field where conversations would be stored so that the user can see their previous conversations.
Thank you very much and congratulations on the great work!
]]>Hi, I′m using 2 custom assistants on my website, everything works fine when I am not logged in, as soon as the users log in, the chats become generic.
This happens regardless of the settings I have in the plugin: All , logged-in, visitors
is there a way to fix this?
Thanks, this plugin is great. It would be great to be able to configure each chat separately via interface or parameters in the shortcode (Chatbot Settings, Prompts and Greetings and Additional Settings)
Best Regards!
]]>Hi, I’ve just setup and customize the Kognetiks plugin for my website. Although the API key has been loaded correctly, the “Message tab” of Kognetiks shows the following: “API Error Type: invalid_request_error Message: You have insufficient permissions for this operation. Missing scopes: model.request. Check that you have the correct role in your organization (Reader, Writer, Owner) and project (Member, Owner), and if you’re using a restricted API key, that it has the necessary scopes.”.
Is there anything I need to customize on the API/ChatGpt side to make it work ? (i.e. is there something other than the API key to setup ? ). Any hint welcome !
]]>I have set up my OpenAI API key and the GPT assistant ID (asst_******************). The chatbot is only working in ChatGPT generic mode but the GPT assistant ID is ignored. How to solve this problem? Thank You!
]]>Great plugin! Can you make the plugin export to PDF? I think it would make it so much better.
]]>I’m getting this message off and on (mostly on now) since last night: It looks like there could be a technical problem with the OpenAI API. Feel free to try again in a bit to see if things are working smoothly. I check OpenAI API status, which all is green for days and I’ve verified there are funds attached to my API key. This seems to happens no matter the model I use as well.
]]>Thanks for this plugin, will there be a feature where you can upload a document to train the chatbot? thanks
]]>I would like to have the option to dynamically process the assistant function if it requires certain action.
Currently I have forked your package and added this code inside the chatbot-call-assistant.php
file
if ('requires_action' === $status) {
if (function_exists('chatgpt_assistant_function_handle')) {
call_user_func('chatgpt_assistant_function_handle',
[
'apiKey' => $api_key,
'threadId' => $thread_id,
'runId' => $runId,
],
$responseArray
);
}
}
I have my custom plugin for parsing the calling the function itself. You are activly developing this plugin that is why I do not want to fork it any longer but propose my ideas and provide them to you
regards
Hi, the first time you visit the page after clearing cash and using a new inkognito, it loads the wrong avatar for the chatbot. It does this every time you clear everything. But the correct avatar is showing when you refresh the page.
On mobile phones it always loadas the wrong avatar, I dont know how to make the chosen or custom avatars to work on mobile phones.
]]>Hi, thank you for a greater plugin! I just noticed that in some browsers, for instance the browser that Android mobile phones uses when I click on the link in a Facebook Messenger chat, the shortcode for loading a custom floating assistant is not working. Instead of loading my assistant, it is just loading a general chatgpt ths not has my specific instructions.
This is strange. Om my laptop, computer, in Chrome browser etc the correct chatbot is loaded. But as ssom as I share the bot in a Messenger chat, if people click st the link from their phones, it loads the wrong bot.
]]>A closing </div>
tag is missing for the <div id="chatbot-chatgpt" style="display: flex;" class="embedded-style chatbot-full">
element.
Love the updates – bullet pointed recommendations working great!
]]>Similar plugins like Chatpress and Themeisle Hyve directly integrate with Assistant for data synchronization. This make it easier to process the data withing the plugin dashboard itself and also give the ability to link required pages or posts. Can you possibly go through similar functionality?
]]>Hi, I’ve recently installed the plugin and embedded it successfuly on my website. The chatbots primary use will be to interact with visitors based on the trained assistant. However, I’ve run into an issue. Whilst the assistant works perfectly in terms of its trained outputs in the assistant page. Through the plug in it eventually time outs about half way through an interaction with an error message that says:
“Error: timeoutOops! This request timed out. Please try again”.
I checked the API Test Results and everything is connected fine.
API STATUS: Success: Connection to the OpenAI API was successful!
I’ve got enough tokens so its not that. I’ve also tried running the plug in with the maximum timeout and am using GPT4 Turbo as I though other versions might be slowing it down.
I’m not sure what left to try to stop it timing out? Any help would be greatly appreciated!!
]]>Hello,
I have a site in two languages, and I was wondering how to display the text “Howdy!!! Great to see you today! How can I help you?”, the personalized greeting “Hello! How can I help you today?”, etc.? I’m using WPML but can’t see anything for it. Do you think it’s possible with another method? Thanks.
Pierre
]]>Hello. I have a blog containing over 100 articles, in two languages, and in two categories. I’d like to use this plugin as a conversational search tool, and have it return a list of links to the corresponding articles rather than answering the user’s question directly. Is this possible?
Thanks
I’m using this plugin with OpenAI assistant. And I’ve attached 5 word files. In the files, there is the statement about how to access a clinick.
If I use Playground with this assistant, it works well. But from Chatbot dialog which is started from the ChatGPT chatbot aikon, it doesn’t work. Just say following messages.
“We are sorry, but the file you uploaded did not contain information about Kochi Vein Care Clinic. Additionally, access to external websites is restricted, so it is currently not possible to search for information directly on the Internet.”
How do I avoid this.
]]>what about streaming responses???
]]>Love the plugin! I’ve been looking for something for a while and this fits the bill (almost haha).
First a feature request – Before I install it on a site that gets reasonable traffic, have we got a way to limit a person’s interactions? Most people are good, but I can see malicious people just hammering away burning up my OpenAI credit for the heck of it. Maybe 10 interactions a session or some other way to limit access?
Secondly, some styling issues. I’ve put the embedded bot in a generatepress widget on the sidebar, and I lose the name of the bot, as well as the additional custom buttons. It might have something to do with that I put the injection code in a div class and have this custom CSS (or it might not):
.custom-chatbot {
font-size: 16px; /* Change the font size to your preference */
}
Finally, would love to include more internal links rather than the one at the moment. Perhaps this could be between 1 and 3?
Thanks for your time – appreciate what you’ve done here
]]>Is there a way to add custom post type in knowledge Navigator??
]]>Hello,
Thanks for your plugin ??
I’m trying to use it in GPT wizard mode where I specially have a file that summarizes all my data. It works well in the OpenAI playground. However, when I test it on my site (same GPT3.5-turbo template on OpenAI and on WordPress), I can’t get the data from my file for simple questions like: where can I contact you or what is your company address?
It works perfectly in the OpenAI playground. So your plugin doesn’t work with files in Retrieval mode?
Can you get back to me or tell me how to do it? ??
]]>