In the web configuration for my assistants, I can select and publish first message mode = assistant speaks first, but the setting does not stick. When I switch to settings for another assistant, then back to the assistant of interest, first message mode is again set to something other than assistant speaks first. Specifically, whichever of the previous options had been most recently been selected. I notice some consistently odd behavior associated with this: 1. When I select "assistant speaks first," I click publish once, it doesn't seem to go through, and then I need to click publish again. With the other options, I only need to click publish once. 2. When attempting to change to assistant speaks first, the drop-down part of the publish button -> view changes shows firstMessageMode"assistant-speaks-first-with-model-generated-message"null on the first click of publish, then firstMessageMode"assistant-waits-for-user" or the other undesired option on the second attempt. There have been a few times I've been able to make assistant speaks first stick. But I can't tell if this is due to the sequence I've messed around with things in or something else.

Hi there, Thank you for your message. Our team is currently out of the office. We operate Monday through Friday, from 9:00 AM to 8:00 PM Pacific Standard Time (PST). We’ll get back to you as soon as possible during our normal business hours. If your message is urgent, please mark it accordingly or include “URGENT” in the subject line, and we’ll do our best to respond promptly. Warm regards, Vapi Customer Support Team

I’ve worked with similar VAPI quirks where “assistant speaks first” doesn’t persist. This usually relates to how the publish action registers changes for that specific mode. I can help you identify why it sometimes requires multiple clicks and why it reverts when switching assistants. We can also review the firstMessageMode behavior in the console to make it stick reliably. Are you seeing this across multiple assistants or just one? @Keith

Multiple assistants.

I replied to you already in my dm

I think I fixed it through the console. I will reach out again if it doesn't stick. Thank you for your help. Solution for others: const token = "xxxx"; const assistantIds = [ "xxxx", ]; const body = { firstMessageMode: "assistant-speaks-first", firstMessage: "xxxxt?" }; (async () => { for (const id of assistantIds) { try { const res = await fetch(https://api.vapi.ai/assistant/${id}, { method: "PATCH", headers: { "Content-Type": "application/json", "Authorization": Bearer ${token} }, body: JSON.stringify(body) }); const data = await res.json().catch(() => ({})); console.log([PATCH ${id}] status=${res.status}, data); } catch (e) { console.log([PATCH ${id}] error, e); } } })();