After we upgraded the WhatsApp integration to v4.5.3, our bot stopped responding. In WhatsApp, the messages show a "read" status (blue check marks), but the conversation is not visible in the Botpress dashboard. The issue occurs with both OAuth and Sandbox connections. Our other bot running v4.2.1 works fine. https://cdn.discordapp.com/attachments/1425549610814013612/1425554721594146987/Screenshot_2025-10-08_214200.png?ex=68e802cb&is=68e6b14b&hm=d6fad55adb6d461b61cc1d86d780fdfa079ab744b3eb96a5b7b20782a87681b1&

Hi Botpress Team, Since updating to Webchat v3.3, I’ve noticed that the

TypingIndicator

appears to be always active for every chatbot using this version. This happens regardless of whether the bot is actually “typing” or not. In the v3.2 update, the TypingIndicator behaved as expected, it only appeared while the bot was processing or sending a message. However, after the update, the indicator remains visible at all times, which can be confusing for end users. Unfortunately, we cannot switch to 3.2 because our customer has seen this Proactive Message and reeeaally wants to keep it 🥲 Does anyone else have this problem? Best regards, sq2-ls https://cdn.discordapp.com/attachments/1425507426421047336/1425507426613858354/image.png?ex=68e7d6bf&is=68e6853f&hm=2c4c2b39e567220f4a2f2c4d05c22ba4d11656c7948a605957dd1e6a8a0eab4a&

a user used the chatbot on the website and it somehow got routed back to an old version of the chatbot the was live back in July. It uses the welcome message used back in july, the old workflow for tracking, and the responses look they are being generated by gpt4o instead of Gemini flash. All of these behaviors are exclusive to the july version. Do you know how this could have happened. The conv id is conv_01K04JD1D4ETGGWPNZXXR34G9Z

Hi guys. Can you tell me why this code to delete the conversation pasted in my flow in execute code doesn't work? https://cdn.discordapp.com/attachments/1425479984452993155/1425479984679620891/IMG_8117.jpg?ex=68e7bd30&is=68e66bb0&hm=e87b5beecdd551158770383767dc4394b06c2ba3520ca48a7f1d8465509c5578&

I want to create a website customer support chatbot that can answer queries about products. This requires the crawl of a c.3500 page website. I tried doing this on the free plan and it only managed to crawl 100 pages then errored out, even when I tried to add more URLs manually. It says the VectorDB usage is 51MB/100MB. Does this sound right? It means I'd need the Team plan to get 2GB of VectorDB storage just to cover all the web pages. Other bot builders I've tried seem to handle this amount of pages in their Knowledge Base fine (e.g. Chatbase). Am I missing something in how to crawl the pages (they are fairly short, we're not talking about thousands of words on each page!) I also see that I've used over $4 of the free $5 AI credits, even though I haven't built a bot yet. I assume this is the cost of AI training on the pages that have been indexed - but again, over $4 for 100 pages seems a very high cost and would mean $140 just to train on the website crawl. I need the crawl updated at least once a day as details change (particularly stock), so that would be a crazy monthly cost! Am I right with my calculations or have I fundamentally misunderstood something and there's a much more efficient way to do this?

Hello everyone. I want to personalize the message that appears when the bot asks: 'Are you sure you want to cancel? Yes/No.' Do you know how to do it?

#Objective Filter from which ad someone answers from to initiate a conversation workflow #Issue I don't know where to start and how to get such information using botpress #Solutions I have asked chatgpt and apparently there's a "ad_id" and a way to track it... but I don't really know where to start and have confused myself with enough AI and Meta/Botpress documentation #Context I'm trying to get which ad someone writes from to give my client better analytics

Is it possible to change the internal code of the autonomous node so that, instead of performing an action listen and waiting for the user’s input, it finishes its execution?

Whenever the bot is trying to show a video or image, it seems like the playback does not work. It has the link to send which gets it right, but the player does not work, any ideas on how to fix this? https://cdn.discordapp.com/attachments/1425103403599794247/1425103404019093565/Screenshot_2025-10-07_at_14.51.02.png?ex=68e65e78&is=68e50cf8&hm=58b06ed87357d49d1a2ed6c045bdc145407e9f6d7f2cb0004f827593ca20d7b4&

I have trouble with exporting my bot, it shows

Export Error: There was an error during the export of your bot:Network Error

I tried multiple times, hard refresh, loggin out and logging in again and nothing worked.

Hi Botpress Community, I’m using an Autonomous Node in my flow to chat with customers about orders. If a customer mentions an item but doesn’t give details (like name or number) to finish the order, I want the bot to move to another node after a timeout (e.g., 1 minute) if they stop replying. I can’t find where to set this timeout. Can someone tell me how to add it in simple steps? Thanks!

Hi, I have deployed Botpress in Docker following the instructions here: https://v12.botpress.com/. Now, I want to add a node in my chatbot that can call my locally hosted LLM. Is it only possible to do this via an API? If so, how can I configure it to send different content in the request body each time this node is called? Thank you!

what do you need? Please tell us more.

Hello community. By default, how long does Botpress keep the conversation, the phone number and the name collected from a WhatsApp conversation? I need it to stay in good standing with the GDPR

I've spend 100$ testing this feature crawling 2 sites with more than 8000 pages. After crawling it, in several batches, the indexed pages dissapear... I've re-added them now for the 3rd time.... How recrawl works? The sitemap.xml files are OK. No robots.txt needed. I'm very dissapointed with this feature, or Is it my fault? Please, can you have a look at https://studio.botpress.cloud/ed68b81a-c51b-4c14-ad75-e93f3d9b626d

Hi, I need to verify my email but my domain has changed so the email is slightly different. Can I update my email?

I want to send web view links to the user on WhatsApp (through the integration), which when clicked by the user opens up the site on WebView (Android or iOS). How can I do this?

Hi; i have client who is using the Shrine theme on shopify, when we try to integrate the bot, the entire website gets zoomed in. We have tried almost everything we can and nothing makes a difference. If someone has any experience, i would love to know how you fixed it. Thanks

Hi everyone! 👋 I’m using the PAYG (Pay-As-You-Go) plan and I’m a bit confused about the message/event limits. On the official pricing page, it says 500 messages per hour, but sometimes I see references to 500 per month. Also, my dashboard hasn’t updated the usage counter for the last 4 days. Could someone clarify: is the limit really per hour or per month? I want to make sure I’m calculating correctly for my clients. Thanks in advance for any help! 🙏

I am having problems using a bot because when I move between different workflows, the screen goes black and does not display the nodes. I reload the page, but it still does not work. If anyone knows how to fix this, please let me know.

It seems no matter what I do when I am using an Autonomous Node to answer a user's queries, and even though we have our website indexed daily, with over 960 pages, every time a question is asked, it calls for a global search and will not access the Knowledge base. I get the same message over and over. The built-in Knowledge base Optimizer is explicitly disabling the KB lookup up. Has anyone else had these issues? It just seems to enact global search no matter what. Even though I have put in the autonomous node instructions not to do so

Hi bot-makers! I am currently building an agent that is deployed on various DM platforms (Instagram and WhatsApp). I am experiencing a "fenomenon" that is common in the way of communication on social media platforms. That is that people often split messages into multiple messages. Example: User: Hi! User: What is the shipping time? By default, this will make the Autonomous Node answer the two messages seperately, even though it is intended to be one message, because it was sent directly after. My question is... Has anyone done a work-around or solved this problem? Or does anyone have any ideas what I can do? Thanks, Lowe

I am testing botpress for a voice support. The text/chat support the studio / flow & output was good. I assumed integration with twilio will be seamless with less change. But when I integrated twilio and tested with a call, all I got was a Upon deep diving I noticed that the error is thrown when a singlechoice has been introduced in the flow error: zs [Error]: An unexpected error occurred in the integration. Bot owners: Check logs for more informations. Integration owners: throw a RuntimeError to return a custom error message instead. at bC (/var/task/customer_code.js:14:18732) at nt (/var/task/customer_code.js:14:18551) at xC (/var/task/customer_code.js:14:19121) at /var/task/customer_code.js:14:169829 at process.processTicksAndRejections (node:internal/process/task_queues:95:5) at async flushSendMessages (/var/task/customer_code.js:3172:8403) at async Promise.allSettled (index 2) at async Object.flushSendMessages (/var/task/customer_code.js:3172:10165) at async /var/task/customer_code.js:3371:2594 { isApiError: true, code: 400, description: 'An error happened during the execution of a runtime (bot or integration).', type: 'Runtime', error: undefined, need help fixing this? If it doesnt work, I will have to cancel subscription and stop work on botpress. thanks 😩

Hello, I’ve been using Botpress for almost two years, and this is the first time I’ve encountered this issue. When I click Publish, nothing happens. The logs show no changes. What I’ve tried • Edited code and flows — behaviour doesn’t change. • Reverted recent changes and removed nodes. • Checked the dashboard: no errors, but the version never updates. • Im deploying to WhatsApp, so it shouldn’t be a browser cache issue. My last successful publish was almost two hours ago. Since then, none of the subsequent versions have gone through. I’m stuck on the same version and urgently need to update it — I have a client waiting, and an event with thousands of attendees relying on these changes. Is anyone else experiencing this, or is there a known workaround? Thanks. just in case someone needs my bot ID 9754e004-3e95-498a-a76d-d3f928d794d8 latests botpush event id evt_01K6K023KQ2DHFZ94VRE6MYH87 createdAt 2025-10-02T17:38:31.159Z type botpublished status processed failureReason null

I have an autonomouse node with an iteration to a workflow card. For testing i cloned my workflow and used it https://cdn.discordapp.com/attachments/1423324171882070027/1423324172372676648/image.png?ex=68dfe56e&is=68de93ee&hm=580ed9410ea308cea4e4454e296c9fe75190a6ef54ac31669467f759927fc39c& https://cdn.discordapp.com/attachments/1423324171882070027/1423324172792102965/image.png?ex=68dfe56e&is=68de93ee&hm=9a58f533e802f09452f84950a574f451ed441a69f6c6d396dc7e54a03482ab80&

I had a talk with my client today and when we discussed about the costs, he asked me about cybersecurity and costs in general. He made a point that some people may use the robot with evil intent to waste company's money. So im asking for a feature that limits AI spend for processes. For example, maxium amount a bot can use for its processes is $0,10 per message. When a user asks a question, the bot scans /analyses predicted cost. And if that process price limit is exceeded, it guides the conversation in different direction or choses a different approach to process that task.

Hello, I’m working on implementing conversation tracking by upserting data into a table so I can later export it. My goal is to store the transcript, summary, user email, as well as the node, card, and workflow information. However, I’ve noticed that when I save the node, card, and workflow, I only get their IDs. Is there a way to capture and store their names instead of just the IDs? https://cdn.discordapp.com/attachments/1423232439194488844/1423232439521771520/Screenshot_2025-10-02_105634.png?ex=68df8fff&is=68de3e7f&hm=1bfe6792e8a771c6dbb532777c9cb261d210f50f31695194c1adbcc3cab54c36&

https://aistudio.google.com/status Tho in their server status, doesn't have any problems? Is it maybe Botpress side? I tested various Models, gemini seems the best use-case for my bot. Is anyone having the same issue? Can this be solved? It would be very bad if the bot was posted to production, and this keeps happening. https://cdn.discordapp.com/attachments/1423231832882942032/1423231833398710443/gemini_outage.jpeg?ex=68df8f6f&is=68de3def&hm=42819c2332f041d6b6f90512ad99bdcdceefa66325942e47389415fd65ace00a&

Hi, I need my bot to accumulate multiple user messages (when users send 2–3 inputs quickly) and send them together for one single consolidated response. Right now, Botpress replies to each message separately. 🙏