Support

Support is a built-in request (ticket) system for contacting the AI Textura team. Here you can ask a question, report a problem, or get help with setup — and receive a reply right inside the platform, without switching to email or messengers. Besides regular tickets, you have two “fast lanes” to support:

The thumbs-down button on any AI reply — starts a 4-step review that creates a ticket automatically.
The Support access to account toggle — lets the team look at your chats, agents, and settings so they can solve the problem faster.

Account Settings

Your account preferences

Billing

Balance and charge history

AI Employees

Managing AI Employees

Channels

Connecting communication channels

You’ll find this section in the left menu: Support.

Support access to your account

At the top of the Support page is the Allow support to access account toggle (new in v2.5.0). When it’s on, the status Access granted appears next to it.

State	What it means
Access granted	The AI Textura support team can sign in to your account on your behalf to look at chats, agent settings, talents, the Knowledge Base, and other sections — to diagnose the problem
Access denied	Support sees only your ticket text and attached files — with no way to sign in to your account

If your ticket is about agent setup, AI behavior in a specific chat, or connecting a channel, turn on access before sending the ticket. This speeds things up considerably: the team doesn’t need to ask you for screenshots and videos — they’ll look for themselves.

Access granted stays in effect until you turn the toggle off yourself. Once the problem is solved, we recommend turning access off again. On critical accounts with financial operations, turn on access only while a specific ticket is being investigated.

How to report an AI reply (thumbs down)

If an AI Employee gave a customer a wrong reply, you don’t need to create a ticket manually — the message has a Mark reply as unhelpful button (thumbs down). It starts a 4-step review: the platform first tries to explain the cause of the bad reply on its own, and if you still have questions, it creates a request to the AI training team with an automatically attached “Context report”.

Video guide

Where to find the button

In the Chats section, open any conversation. Every AI Employee message has a thumbs-down button (aria-label: Mark reply as unhelpful). Click it to open a review dialog.

Step 1 of 4 — Bad reply review

Step 1 — automatic diagnostics of a bad reply

The platform shows automatic diagnostics of how the agent arrived at this reply:

Agent decision step	What it shows
Language detection	Which language was detected in the customer’s message (for example, “Message language: English”)
Tools used	An expandable list of talents the agent called (`check_language`, `crm_create_lead`, etc.)
What the agent did	The source of the reply: general knowledge, the Knowledge Base, a talent, or a combination
Final decision	The resulting action — for example, “Nothing was sent to the customer — the agent stayed silent (strict mode with no relevant sources)”

At the bottom are three buttons:

Button	When to choose it
Cancel	Close the review without sending
It’s the settings	If the diagnostics already show the cause is in the instructions or prompt — go to step 2 to check
Something else →	If the diagnostics don’t explain the problem — skip the check and describe it in your own words (step 3)

If automatic diagnostics aren’t available, the platform shows: “Couldn’t get automatic diagnostics. Review the timeline below manually.” This is rare — proceed as usual and describe the problem in step 3.

Step 2 of 4 — Check the instructions and tools

Step 2 — checking instructions and tools

“The instruction may be incomplete — the agent didn’t understand it needed to call a tool.” At this step, the platform shows the list of tools the agent used in this case (check_language, Knowledge Base search, CRM operations, etc.). You decide whether the Persona or the working rules themselves are to blame:

Button	What it means
← Back	Return to the automatic diagnostics
In the instructions	Confirm that the cause is in the agent’s instructions — go to step 3 to report to the training team
Instructions are OK →	The instructions look fine — describe the problem separately (step 3)

Before clicking In the instructions, open the AI Employee’s Persona in a separate tab and check whether it actually describes how to call the tool in question. Sometimes fixing the prompt is faster than waiting for a support reply.

Step 3 of 4 — Describe the problem

Here you fill in three fields — they make up the body of the ticket:

Field	What to enter
What should the reply have said? (textarea, required)	Describe the correct reply or what was missing from the actual one
What do you think caused the error? (select)	Choose one of the categories — see the table below
Additional comments (textarea)	Any useful information for the training team

Cause options

Cause	When to choose it
Inaccuracy (made-up facts)	The AI gave information that doesn’t exist in reality or in the Knowledge Base
Misunderstood the question	The AI answered the “wrong” question — it misinterpreted the customer
Didn’t use knowledge / tool	The agent has a Knowledge Base or talent but didn’t use it
Rudeness / inappropriate tone	A style problem — the tone or ethics were violated
Too long / too short	The reply format didn’t fit
Other	Any other situation — explain in the text fields

The Next → button becomes active only after you fill in the first textarea.

Step 4 of 4 — Thank you! Your feedback matters

The final screen gathers everything into one ticket:

Summary: Expected reply: {your text from step 3}, Cause: {selected cause}.
Message: “This information will go to the AI training team. We use it to make agents better. A reply usually arrives within 3 business days.”
Key point: “The case context report will be automatically attached to the support request.”
The 📎 Open context report → link — lets you view a detailed report of the agent’s decision steps before sending.
Buttons: ← Back, Send.

After you click Send, a new ticket with the On Customer status appears in the Support section — the AI training team usually gets back to you within 3 business days.

Thumbs down is not just a rating for analytics. It’s a ticket format for a specialized team: it automatically collects all the technical information (message language, talents used, Knowledge Base search results, the agent’s final decision) and attaches it to the request. The team doesn’t need to ask you for context — it’s already there.

Use thumbs down for any inaccuracy in an AI reply — even if you’re not sure the problem is serious. The more of these signals the training team gets, the faster agents become more accurate.

How to create a new ticket

Open the Support section

Click Support in the platform’s left menu.

Click + New ticket

The button is at the top of the page.

Enter a title

Briefly describe the essence of the question or problem.

Write the first message

Explain in detail what happened or what you need. Specify which AI Employee or feature it concerns, what you did, and what went wrong.

Send the ticket

Once created, the ticket appears in the request list with the On Customer status.

The more precisely you describe the problem in your first message, the faster you’ll get a reply. If the problem is about a specific AI reply, try the thumbs-down button first: it attaches the context report automatically.

Messaging inside a ticket

After you create a ticket, the conversation takes place in chat format — right in the right-hand panel of the Support section. Your messages appear on the right, the team’s replies on the left. To send a message:

Open the ticket by clicking it in the list.
Type your text in the input field at the bottom.
Send the message.

When the team replies, a red indicator with the number of unread messages appears next to the ticket in the list.

If you want to add to your request, just write a new message in the same ticket. You don’t need to create a separate request.

Ticket statuses

Each ticket has one of two statuses:

Status	What it means	What to do
On Customer	Waiting for your reply or just created	Check whether the team has a question for you — and answer it
Done	The request is closed, the issue is resolved	Nothing — the ticket is archived

The status updates automatically: when the team replies and waits for your response, the ticket moves to On Customer. When the issue is resolved, the team moves the ticket to Done.

A closed ticket can’t be reopened. If the problem happens again or you have a new question, create a separate ticket. You can reference the previous request in the text.

Searching and filtering requests

If you’ve accumulated a lot of tickets, two tools help you find the one you need quickly:

Tool	Purpose
Search requests	Text search across ticket titles
Status filter	Show all, only On Customer, or only Done

Response times

The AI Textura team tries to respond to requests as quickly as possible. Approximate times:

Request type	Approximate response time
Critical problem (platform down, charges)	Within a few hours
Question about setup or features	Within one business day
Suggestion or feature request	Within 1–2 business days

If your question is urgent, say so in the ticket title. For example: “URGENT: AI Employee stopped replying in Telegram”. This helps the team prioritize your request.

Tips for effective requests

One question — one ticket. This makes it easier for the team to track and resolve each task.
Answer the team’s questions — it speeds up the resolution.

To get the fastest response to your request, attach the following to your ticket:

A screenshot of the error — attach the image right in the message so the team can see what’s happening immediately.
A link to the conversation — open the Chats section, find the conversation, and copy the URL from your browser’s address bar. That way support can see the conversation and understand the situation.
Steps to reproduce (if the error is complex) — describe step by step what to do to reproduce the error. This lets the team repeat the scenario and see in the conversation how it works.

Example of a good request:

Title: “Booking link doesn’t work” Message: “The ‘Concierge’ AI Employee in Telegram sends customers a booking link, but it opens a blank page. Link: [URL]. The problem started yesterday. Conversation link: https://app.aitextura.com/chats/… To reproduce: write to the bot ‘book a room for March 20’ — the reply will contain a broken link.”

FAQ

Can I attach a screenshot or file to a ticket?

Yes! You can attach images and files right in a ticket message. We recommend always adding a screenshot of the error — it speeds up the resolution considerably.

How do I know I've received a reply?

A red indicator with the number of unread messages appears next to the ticket in the request list.

Can I reopen a closed ticket?

No. If the issue comes back, create a new ticket. You can reference the previous request in the text.

How many tickets can I create?

There’s no limit. Create as many requests as you need.

Who answers the requests?

The AI Textura support team — real people who know the platform inside out. It’s not a bot.

Is support available 24/7?

The team works during business hours (Moscow time), but critical requests may be handled outside the standard schedule.

How is thumbs down different from a regular ticket?

Thumbs down is a specialized channel for requests about AI reply quality. The platform collects all the technical information itself (message language, talents, Knowledge Base, final decision) and attaches it to the ticket. The request goes to the AI training team. A regular ticket is a free-form request to general support: questions about billing, connecting channels, access, and so on.

Is it safe to turn on Support access to account?

Yes. Access is granted only to AI Textura support staff and is used to diagnose your problems. All actions are logged on the platform side. Once the ticket is resolved, we recommend turning the toggle off — you can do this at any time on the Support page.

What's in the Context report attached via thumbs down?

The report includes: the language of the customer’s message, the list of talents called and their results, what was found in the Knowledge Base, and what final decision the agent made. You can view the report before sending via the ”📎 Open context report” link in step 4.

The platform or my account won't open

First, check status.aitextura.com — maintenance and outages are posted there, and urgent updates are duplicated in the platform’s Telegram channel. If the status is green, try another browser and clear your cache, then write to us.

The AI Employee stopped replying completely -- where do I start

In order: your word balance isn’t at zero; the AI Employee is turned on and its channel is active; turn the channel off and on again — this rebuilds the connection; check status.aitextura.com. If everything is fine, create a ticket and attach a link to the conversation.

OpenAI is down -- what should we do

Nothing. The platform switches to backup models automatically; replies may be slower than usual during that time.

Which models does the platform run on

On models from OpenAI, Google, and other leading providers — the platform doesn’t train its own models. The platform itself is written in NodeJS with Python modules.

Where can I watch training videos

On the AI TEXTURA YouTube channel. Videos for specific sections are embedded right in the documentation articles — on the Persona, Knowledge Base, Talents, Notifications pages, and others.

Getting Started

AI Employee

Talents

Channels

Conversations

Employee Settings

Account

Account Settings

Billing

AI Employees

Channels

Support access to your account