Give every customer a fast, useful answer. Text AI Agent resolves routine requests, recommends the right products, captures leads, and knows when to bring in a human.
14-day free trial No credit card required Easy setup
30,000+ teams already turning conversations into revenue
REVENUE FROM SUPPORT
An AI customer service agent is software that can understand a customer's request, answer from approved business knowledge, and take the next useful action. Unlike a scripted bot, it does not need every conversation mapped in advance.
Text AI Agent can answer product and policy questions, collect contact details, create or update tickets, recommend products, and hand the conversation to a person when that is the better path.
Build your first AI agent ›
![]()
“The fact that no IT support is needed to make this work is incredible. We’re able to efficiently reduce tickets and drive sales ourselves, without having to wait for the next sprint.”
Duncan Andrews, CRM & Digital Manager, Experiences by Wembley Stadium
FROM SETUP TO REPORTING
Start with what your business already knows, then decide how the AI should speak, act, and report results.
Add public website pages, help center content, product details, policies, individual URLs, editable articles, or supported files. Assign the right sources to each AI agent.
Choose a tone, answer length, and languages, then add the skills your agent needs: welcome visitors, collect details, create tickets, transfer conversations, or build Custom Skills.
Decide where an AI agent should work by website, page URL, or visitor location. Separate teams, brands, or use cases when your operation needs different knowledge and actions.
Track handled chats, transfers, resolutions, resolution rate, and customer satisfaction. Shopify teams can also connect conversations with sales during the chat and for up to 30 days.
An AI resolution counts only when the AI directly solves at least one customer question. Tests, spam, greetings, and transfers without an answer do not count.
CONTROL WHAT IT SAYS
Choose the sources, keep them organized, and preview how they shape a response.
Define when a conversation should reach your team, with nothing lost in the move.
![]()
“Our goal was to decrease inbound phone calls while retaining or improving total customer interactions. We realized and surpassed these goals within the first month of implementing LiveChat.”
Kyle Therriault, VP of Internet Strategy & Business Development, Auto Accessories Garage
ANSWERS, NOT DEFLECTION
Customers should not have to fight a bot to get help. Text AI Agent uses the knowledge and actions you approve, gives a direct answer when it can, and brings in a person when that is the better path.
Answer recurring questions
About products, delivery, returns, billing, and policies.
Use connected skills
Check details, collect information, and update a ticket.
Read sent images
When a visual detail helps explain the issue.
Keep service available
Outside your team's working hours.
NOTHING LOST IN TRANSIT
Keep the customer history together when a conversation moves between AI, a person, web chat, email, Messenger, WhatsApp, SMS with Twilio, or a follow-up ticket.
SERVICE MEETS SALES
A service conversation often begins the moment a customer is deciding what to buy. Text helps your team respond to that intent while it is still active.
Use the customer's question and browsing context to guide them toward a useful choice.
Try product recommendations ›
With Shopify, customers can choose product variants and add items to the cart inside the conversation.
Try in-chat checkout ›
Collect contact details and the information your sales team needs before the handoff.
Try lead capture ›
Use the AI Agent Sales report and Sold view to connect conversations with revenue.
See revenue reporting ›
![]()
“We could give our customers the chance to rate our consultants. More importantly, we can monitor the quality of customer service thanks to advanced analytics.”
Milena Wojewoda, Digital Project Specialist at Sephora
BEYOND ANSWERS
Answers solve only part of the customer's job. Custom Skills let an AI agent use live context, exchange data with your systems, and complete a defined action without sending the customer elsewhere.
Pass customer details, page context, and information collected during the conversation into a skill.
See live context in a skill ›
Use webhooks and APIs to look up data, update a record, or trigger your existing workflow.
Connect your first action ›Describe the inputs and output the AI should use without building a rigid conversation tree.
Build without code ›
Use the Runs view to inspect how a Custom Skill performed and improve it over time.
Review a skill run ›WHERE TO START
Choose a frequent customer need with a clear outcome. Prove the value, then give the AI agent more knowledge and skills.
Help customers compare options, understand product details, and find information while they are deciding.
Resolve common delivery, return, account, and policy questions before they become manual work.
Keep a useful first line of service available when your human team is offline or at capacity.
Collect the right details, classify the need, create a ticket, and send it to the right team.
Start a relevant conversation on selected pages when a visitor may need help moving forward.
Guide shoppers to the right product and collect the details your sales team needs.
BUILT AROUND YOUR TEAM
Start with one focused use case, connect it to the team and systems you already have, and expand when the outcome is clear.
On Growth and Enterprise plans, create multiple AI agents with separate profiles, knowledge, skills, languages, and teams.
Connect tools such as Shopify, Salesforce, HubSpot, Slack, Mailchimp, WordPress, and OpenAI, or use MCP, webhooks, and APIs.
Monitor conversations, transfers, tickets, customer satisfaction, AI resolutions, and sales results without splitting the operation across products.
![]()
“Customers love LiveChat and instant answers from a local specialist. They’re often on a tight schedule, so LiveChat lets them book with immediate confirmation.”
Leah Vigil, Director of Marketing, Luxury Estates International
TRUST AND CONTROL
Clear control over what each AI agent knows and what it can do, with enforced two-factor authentication and a full audit log behind it.
SOC 2
GDPR
CCPA
PCI DSS (SAQ A)
Data Privacy Framework
WCAG 2.2
BBB Accredited
LIVE IN DAYS, NOT QUARTERS
You do not need to rebuild your service operation. Start small, test with real questions, and expand from evidence.
Pick a common question, repeated task, or buying moment your team can measure, then connect only the knowledge that role needs.
Define how the AI speaks, what it can do, and when it should transfer. Use preview to check answers before launch.
Track resolutions, satisfaction, transfers, and outcomes, then improve from real conversations.
WHAT CHANGES
Resolve common questions as soon as the customer asks, including outside normal working hours.
Move repetitive work to AI so people can focus on complex and high-value conversations.
Ground answers in the sources, voice, and instructions approved by your business.
Use context to recommend a product, qualify a lead, or complete the next service action.
Keep customer details and conversation history with the request when a person takes over.
Connect AI performance with resolutions, satisfaction, team workload, and attributed sales.
NOT A DECISION TREE
A basic chatbot follows paths you write. An AI customer service agent can interpret an open question, use approved knowledge, and choose a relevant action.
| Capability | Basic chatbot | Text AI Agent |
|---|---|---|
| Conversation logic | Prebuilt paths and buttons | Understands open customer questions |
| Answers | Fixed messages | Uses approved business knowledge and context |
| Actions | Limited scripted steps | Runs built-in and Custom Skills |
| Handoff | Often a separate flow | Transfers in the same inbox with conversation history |
| Improvement | Paths require manual rebuilding | Review answers, skill runs, resolutions, and outcomes |
An AI customer service agent answers questions, completes routine service tasks, recommends relevant products, captures leads, and passes a conversation to a person when human judgment is needed. Text AI Agent uses your business knowledge, conversation context, and connected tools to move each request toward a useful outcome.
Already using LiveChat? If you are on a LiveChat plan rather than Text, AI Agent is not included automatically — see how to move to Text to unlock it.
You control the sources Text AI Agent can use. Add selected website pages, help center content, files, or articles, then preview answers before launch and review conversations after launch. Sources can be assigned to individual AI agents so each one answers only from the knowledge relevant to its role.
Yes. You can choose the languages an AI agent should use and set its tone and answer length. On Growth and Enterprise plans, you can also create multiple AI agents with separate teams, profiles, knowledge, skills, and language settings.
An AI resolution is counted when Text AI Agent directly solves at least one customer question in a conversation. Tests, spam, greetings, and transfers where the AI did not answer a question are not counted. A conversation can count as no more than one AI resolution.
On a legacy LiveChat plan? AI Agent is not part of that subscription, so you would need to move to Text for resolution-based AI Agent access.
AI Agent is a Text-only feature. If you signed up for Text, or have already moved to it, AI Agent is included up to your plan's monthly resolution allowance. If you are still on a legacy LiveChat plan, AI Agent is not part of your subscription.
Compare plans or read more about moving to Text to get AI Agent.
Only what you approve. Choose the pages, files, and articles each AI agent can use, then preview and review its answers before and after launch.
Text meets SOC 2, GDPR, CCPA, PCI DSS (SAQ A), Data Privacy Framework, and WCAG 2.2 standards.
AI agents are best used alongside people. Text AI Agent handles repeatable work and common questions, while your team can monitor active chats, take over at any time, and handle conversations that need judgment, empathy, or an exception.
Yes. You can define when Text AI Agent should transfer a conversation. The customer, conversation history, and collected details stay in the same inbox so a human agent can continue without asking the customer to start again.
You can start by connecting approved knowledge, choosing a voice, adding the skills you need, and testing in preview. A focused first use case can be launched without mapping a long decision tree or rebuilding your service operation.
AI assist features such as response auto-suggests, tag suggestions, text enhancement, and chat summaries are included in every LiveChat and Text plan. AI Agent, the autonomous agent that resolves conversations on its own and is billed by resolution, is a Text-only capability.
There are three Text plans — Essential, Growth, and Enterprise, starting from $19 per month — and each one carries its own monthly AI Agent resolution allowance. Legacy LiveChat customers can move to Text to get AI Agent.
Yes. Text supports integrations and workflows with tools such as Shopify, Salesforce, HubSpot, Slack, Mailchimp, WordPress, and OpenAI. Custom Skills can also use webhooks and APIs to exchange data with your own systems.
You define the skills an AI agent can run and the rules for handing a conversation to a person. You can monitor active conversations and let a human agent take over at any time.
Put Text AI Agent to work
14-day free trial AI Agent that resolves and sells No credit card required