since yesterday i am unable to initiate any outbound call with any assistant. I am getting this: call.start.error-get-transport You can check this call id 019ae38f-47d1-7336-867d-e5fb2b1244ef https://cdn.discordapp.com/attachments/1445745496084709466/1445745496290365502/image.png?ex=693176ea&is=6930256a&hm=3871b0af53999945859fc6a7dab449f5cfb68f5d574db1cd71fc142831130e91&

in all my agents i cant find any more the code snippet - does anyone else have this problem?

It clearly shows that tool calling was successful and provided a result. But the AI keeps gaslighting and telling me that the results haven't come in yet. What could be causing this? Call ID: 019ae3cb-7d41-7cc2-8b84-790b029312bb https://cdn.discordapp.com/attachments/1445727355753594931/1445727356193865728/image.png?ex=69316606&is=69301486&hm=97ecb1b0407d93b099ca242c06c198f191e844ee91144de7715e68009431cf3a&

hi, i'm trying to connect my Rime key to Vapi and it's not accepting. Does Rime integrate with VAPI?

Is there a way to setup evals for different projects that use common evals?

I am writing to inquire about the behavior of outbound calls using the POST /call endpoint when the concurrency limit is exceeded. Based on a recent email announcement, I had expected that Vapi would handle the queuing of calls when the concurrency limit is reached. However, we are receiving an error (e.g., HTTP 429) for requests made on or after the scheduled release date (Nov 21) when the limit is surpassed. Could you please provide support for investigating the cause of this issue? Demand for the service built using Vapi is rapidly increasing, which has created a business-critical situation. https://cdn.discordapp.com/attachments/1445661102162706525/1445661102766555166/Action_required__queued_processing_for_POST__call_and_response_chan__-_mail.google.com_1.png?ex=69312851&is=692fd6d1&hm=ca9c7cfd257ea9c391f9625a200a9413e79ebe3d50c4a6f5bfcd4394ba67f832& https://cdn.discordapp.com/attachments/1445661102162706525/1445661103295172671/2025-12-03_15.13.31.png?ex=69312852&is=692fd6d2&hm=1695c0ca6ce920c8d81aaecdbc126bce3f29d508094de6620d3486accee22427&

I'm experiencing an issue with dynamic variables not being properly rendered in my system prompt after a successful

apiRequest

tool call. I use the dashboard. I've thoroughly debugged the issue and would appreciate your assistance. Issue Details: - I've configured an API request tool that returns data including fields like date, dayOfWeek etc. - I've mapped these responses to dynamic variables in the tool configuration in response body. See attached images. the Response Body is auto generated when I test the tool call using "Save Schema" - When I check the call logs, I can see the tool call executes successfully - However, the system prompt is not being re-rendered with the populated variables (I checked the model Azure OpenAI HTTP request -> attributes.request.messages) - All variable references (e.g., {{data.dayOfWeek}}, {{data}}) are being replaced with empty strings even after the tool call completes Debugging Steps Taken: - Verified the API request tool is executing successfully in the call logs - Confirmed the API is returning the expected data structure - Added debug statements to the system prompt to check variable rendering - Verified that built-in variables like {{now}} are rendering correctly - Checked multiple conversation turns after the tool call completes - Examined the model requests that occur after the tool call in the call logs and latest model request in the call Am I debugging it wrong? What's the best way to debug it? Additional Information: - Assistant ID: bc4d9f7f-01f2-4370-9604-f77f684fe1a1 - Call ID for reference: 019ae217-d360-7000-8095-7dd4bc00798d I would greatly appreciate your help in resolving this issue. Please let me know if you need any additional information. https://cdn.discordapp.com/attachments/1445637070981824542/1445637072676454530/image.png?ex=693111f0&is=692fc070&hm=95a68dc3625e6c789e2d5c8a16484167124d9cff00ae5adf8e1230f4f40df481&

Hello, support team. While I was integrating vapi to GHL, I faced on this error: "GoHighLevelMCPCredential or its authenticationSession not found" I set GHL api key correctly and GHL tools was connected correctly. I tried to follow your guide, but it didn't work. How can I fix this error? Why this error is occured? Please guide me step by step. Thank you. https://cdn.discordapp.com/attachments/1445608255941513287/1445608256218333194/b1be2588-85f0-4bb9-bea4-f7010f5391d5.png?ex=6930f71a&is=692fa59a&hm=de64d73d43d1161ab77e4bfa461847822135d2e478e4fcaaeca7043e1c0a04f8&

I created a custom knowledge base and added it to the assistant, but now every time I do anything in the dashboard, it overrides it. Is there any way to fix this or a better solution here?

Is it possible to send a background message to update the conversation history from a server? Or is that only possible from a client?

I'm seeing user speech transcription + agent speech transcription fed to model (using open ai 5.1) as the messages context. Can we use the model's output directly instead of agent speech transcription which can have errors? For example: in this call https://dashboard.vapi.ai/calls?callId=019ae00b-36d1-7117-b9e5-1a04fbad0033 - User says "So please send the email directly to me at triston at with pace dot com" - Model outputs "I heard you say triston at with pace dot com" - When agent says this^ it gets transcribed to "I heard you say tristin at with pace dot com" (INCORRECT) - Next OpenAI request sends "I heard you say tristin at with pace dot com" Result is that the email is completely wrong which is a common problem for us. Why is Vapi default using transcription of agent speech instead of direct model output? This isn't how providers like LiveKit work. Is there a configuration I can use to change this to only use user speech transcription + model output?

When i clicked on my assistant from assistants section, it is showing Error: Something went wrong. I am attaching screenshot of the screen and console logs. @User @User https://cdn.discordapp.com/attachments/1445479078017896448/1445479078231674921/message.txt?ex=69307ecb&is=692f2d4b&hm=8fef4b4b7bc9e7734555e913aa3bae5733ff7133b0e92cd19e7328c11464b7ca& https://cdn.discordapp.com/attachments/1445479078017896448/1445479078638649344/image.png?ex=69307ecc&is=692f2d4c&hm=58dca3646354cf0b36b34230625c15bb72fd1f8b8c2fb1436a8e3becedfd2a4e&

I’m reaching out because I’ve been facing a persistent issue with my voice agent’s prompt, and I’ve already tried multiple adjustments without success. My agent is designed for debt collection in Brazil, so the language used is Portuguese (Brazil). Here in Brazil, the correct way to pronounce the local currency is “reais”, not “BRL”. However, despite specifying this clearly in the prompt (and even reinforcing that “BRL” should never be spoken), the bot still pronounces it as “B-R-L” frequently. This makes the audio sound unnatural and not aligned with local standards. Could you please guide me on the best way to ensure the bot consistently says “reais” instead of “BRL”? Any configuration or prompt-structure recommendation that could help would be greatly appreciated. Thank you for your attention and support.

I need your help regarding the workflow configuration. I’m trying to adjust several backend settings — specifically: Call Timeout Settings Messaging Stop Speaking Plan However, I’m unable to find any option to configure these settings in the workflow interface, and you mentioned that these configurations are also not available through the API. This leaves me without a way to properly configure the workflow. At the moment, the workflow is: sending incorrect information, not stopping speech when it should, not adding pauses correctly, and not sending the required data to my webhook. The prompt logic and decision-making are perfect, but the backend behaviour of the workflow is very limited compared to the configuration options available in the generative AI assistant prompts (which offer full UI control). Since workflows seem to still be under development and do not provide the same configuration capabilities, I’m unable to properly prepare this for production. I urgently need a solution or guidance on how to configure these settings for the workflow to function as expected. Could you please advise on how I can configure or override these settings? Any support or workaround would be greatly appreciated.

Multiple test suite are failing when am trying to run 7 test calls Here is the test suite id: c6015601-bf48-47e4-b416-ab266c7a015c https://cdn.discordapp.com/attachments/1445467596513083525/1445467596949426176/Screenshot_2025-12-02_at_9.31.47_PM.png?ex=6930741a&is=692f229a&hm=c3369bcb7d211f1b700d3d16049def0cd249cc9ea771f2ce3b6a3c9a7b2c5cbb&

I've been making batch calls on my own platform using VAPI's API but I have the AI sending a text and instructions to end the call after the text is sent but somehow the assistant won't end the call until the person being called acknowledged or says something after the text is being sent. How can we have the call end right after sending the text? Here's a call ID as an example, the AI was instructed to say end the call after the sms is sent but it only did so after I said ok: 019ae00a-d426-7ee9-89f0-efd6192ce64f

Two days with an error as per the screenshot. Sent email support yesterday. Got multiple Assistants and they seem to be ok, but this one does not work, nor can I access it Can I share the ASSistant ID here? https://cdn.discordapp.com/attachments/1445458559516016851/1445458559893770321/image.png?ex=69306bb0&is=692f1a30&hm=e8d3e6ecf19684f74f47e9a3aa52b585f064a4aa3e36077199f25038861e0b06&

I noticed in the Vapi SDK, we can pass in a key called "variableValues" on the Vapi instance instantiation but the value of these keys seem to be obfuscated /modified once Vapi recieves them and tries to use that value in tool calls. This is a major bug for my use case as it allows me to differentiate among user chat histories. I noticed when I told Vapi to pass this variableValue to the tool calls as a param, it would modify slight characters in the string value. These are some failed logs I found on my server when trying to make tool calls with Vapi. As you can see, it slightly obfuscates the string. Failed requests (10:41:50 - 10:42:32): - ❌ Request 1: 2536d197-fc73-497d-8c8a-e6755bfe404 (corrupted: d199 → d197, missing b) - ❌ Request 2: 2536d199-fc73-497d-8c8a-e67555bfe404 (corrupted: bbf → 55b, 4f missing) This is a CONSISTENT issue, and I've isolated it to be only occurring when the voice agent is required to use a variableValue to pass into a tool call. I have a text mode with a local LLM and it does not experience the same issue. Is there a patch to fix this soon or can your team look into this asap? It is a fairly critical bug, as this allows (not only myself but other users too) me to allow the Vapi instance to use tool calls and pass params in which is vital to the functionality of my app. Thank you.

Hi vapi team, when using my agent everything works well until it's time to transfer the call. Agent initiates it but then the call drops

Hello! I'm new to VAPI 1. Created account 2. Added SIP Trunk 3. Added phone number 4. Pressing "Make a call" on phone number page Every call drops to "Customer Busy" , so I don't even get a call on my phone for test Also tried to initiate a call with API. Result the same 🙁 Help please

Hello VAPI Support, I'm experiencing an issue where my custom tool works when tested directly, but doesn't trigger the webhook when called by my assistant during conversations. DETAILS: - Tool ID: 23a97c05-eb36-4551-b152-2907f3a9ef45 - Tool Name: check_availability - Assistant ID: ed1cee38-797d-4e48-9af5-... - Webhook URL: https://imel.app.n8n.cloud/webhook/vapi-receptionist WHAT WORKS: ✅ Testing the tool directly in VAPI Dashboard → webhook receives request ✅ My webhook endpoint (n8n) works perfectly WHAT DOESN'T WORK: ❌ When assistant calls the tool during chat → webhook never receives request ❌ VAPI logs show "No result returned" error ❌ n8n shows no execution when assistant calls tool I've tried: - Removing and re-adding tools to assistant - Recreating the tool - Verifying all configurations - Checking async settings (it's OFF) - Testing with different prompts The assistant calls the tool (I can see it in VAPI chat logs), but the webhook is never triggered. This only happens when the assistant calls it, not when I test the tool directly. Can you help diagnose why the assistant isn't actually sending webhook requests to my endpoint? Thank you!

I just opened my assistant and am getting this, while others are working. Please hurry up my client needs an update. What should I do now? https://cdn.discordapp.com/attachments/1445402855715635331/1445402856080412682/IMG-20251202-WA0061.jpg?ex=693037cf&is=692ee64f&hm=868ab4bc0f8656d65e2415cbbb701294b68266e88fc06bed5577ec76a5872861&

I have tried many many different body, each time the api would return { "statusCode": 500, "message": "Internal server error. Contact support@vapi.ai for help." }

I enabled the video recording feature, and when I export the call object, I see video link in

recording.videoUrl

. But I want to know where can I find this link in Vapi dashboard directly?

Hello, help is needed, i waited last night to set up some agents and there was anoter error, i wait till now and somes issue are still in place . Please assist us we this issue ! Thanks "Error Something went wrong. This incident has been reported. Contact support if the issue persists." "Error: RangeError Message: invalid_argument Timestamp: 2025-12-02T09:21:42.316Z Path: /assistants/7f6bd906-3b22-4ff9-8165-13a4e18be149" https://cdn.discordapp.com/attachments/1445344512686817280/1445344513299320953/image.png?ex=69300179&is=692eaff9&hm=3e390c2698eec2923873334b232e366ac4b023c69e1cdfb80f42a0f3394b51bd&

Hi team, Our organisation is not able to save the credentials for AWS S3. We were able to do it a few days back, but now there is an error. I am attaching the screenshot of the error below. Organisation ID - 4660cb9e-85c7-405d-9b52-26ca3235ee58 Best regards, Jayesh.

hi there vapi support. we are unable to access this Assistant . Please fix this asap https://cdn.discordapp.com/attachments/1445333093689852028/1445333095732609105/image_3.png?ex=692ff6d7&is=692ea557&hm=f4868b5073fbf822865e28b177a193ec8324fa8e6ee8b50f2c303aceffa338e6&

I need help purchasing more credits. I’ve already added my card to file but I keep getting an error message. https://cdn.discordapp.com/attachments/1445179028498813092/1445179028922306612/IMG_0977.png?ex=692f675a&is=692e15da&hm=5dc8ca07dc6d1a8fdc6edb4703254157257aa2428df1a66f5f6192bab9f5926c&

When i click on my assistan i got the message above. And in Technical details for curious humans Error: RangeError Message: invalid_argument Timestamp: 2025-12-01T22:08:27.443Z https://cdn.discordapp.com/attachments/1445173886957256867/1445173888035066006/image.png?ex=692f6290&is=692e1110&hm=03989b5a0c738214956cf1751fc512043655bb48788d87fadd1202986801d33f&

I launched a campaign and it had all my calls queued instead of actually calling them. I tried cancelling but they've been queued for the past 2 hours, just trying to make sure it actually gets cancelled so it doesn't just like randomly call people in the middle of the night or something.