#Guides

Guides are for adding an agent to the app you are building: one task per page, written to be read by you or by your coding agent. Start with Build an AI agent for your app. The reference for every file, command and endpoint is in the documentation, the CLI, the SDK and the API.

Two ways in, and you can use both:

From any chat: add https://vatio.ai/mcp as a connector in Claude, ChatGPT or Cursor, sign in and ask for the agent (Vatio in Claude and ChatGPT).
To build it into the app's code: in the project, run npx @vatio-ai/skills, then /vatio-init (the skills).

#Start here

Build an AI agent for your app: The whole path, for you and for the coding agent doing it: connect, build, try, put it in the app, publish, improve.

#By use case

Bookings and appointments: Let signed-in users see availability, book, move and cancel through your app's own API.
Orders, deliveries and returns: Order status, delivery estimates, changes your rules allow and returns, for each signed-in user.
Support for signed-in users: Answer from your docs, check the user's own account, troubleshoot and open a ticket with what was tried.
Onboarding new users: Help new users finish setup: answer how-to questions and tell them their next step from their own progress.

#Built in an app builder

Lovable apps: Sync your Lovable project to GitHub, run the two steps in Claude Code or Cursor, and keep building in Lovable.
Bolt apps: Connect your Bolt project to GitHub, add the agent from Claude Code or Cursor, and keep Bolt in sync.
Replit apps: Push your Replit app to GitHub, add the agent from Claude Code or Cursor, and pull it back into Replit.
v0 apps: Connect your v0 project to GitHub, add the agent to the Next.js app, and allow your vercel.app address.

#By stack

Next.js apps: The widget in your root layout, a token signed with jose, and tools that call your Route Handlers.
React apps: The widget mounted in an effect; your backend signs the user's token and your API verifies it.
Supabase apps: An Edge Function signs the signed-in user's token, and tools call your Edge Functions to read that user's rows.
Firebase apps: A Cloud Function turns Firebase sign-in into a token Vatio verifies; tools call your HTTPS functions.
Rails apps: The widget in your layout, users signed in with vatio-sdk, and tools that call your controllers.
Django and FastAPI apps: The widget in your base template, and tokens signed and verified with PyJWT.
Laravel apps: The widget in your Blade layout, and tokens signed and verified with firebase/php-jwt.

#Channels and SDKs

An AI chat for your web app or site: One script tag puts the agent in your web app or site, themed to your brand and able to recognize signed-in users.
An AI agent in your iOS app: The same agent inside your iPhone app with the iOS SDK, signed in as the user your app already knows.
An AI agent in your Android app: The same agent inside your Android app with the Kotlin SDK and a Compose screen, signed in as the user your app already knows.
An AI agent in your Flutter app: The same agent in your Flutter app on iOS and Android, signed in as your user.
An AI agent on WhatsApp: Add WhatsApp to your app's agent: test on a shared number, then connect your business number.
An AI agent on Instagram: Answer Instagram DMs with your app's agent: test on a shared account, then connect your professional account.

#How it works

The AI agent runtime: What Vatio runs between a customer's message and your agent's reply: grouping messages, the prompt, the model and your tools, safeguards, delivery and what follows.
Long-running tools: Tools that take minutes, such as recommendations, reports or a job that drives a browser: the agent keeps talking and delivers the result on any channel.
Best practices: Instructions, tools, knowledge, handoff and tests: what works for an agent real users talk to.
Glossary: The words around customer-facing AI agents, from tool and handoff to eval and preview, in plain terms.

#Compare

Choosing how to build your app's agent: An agent framework in your backend, a visual or WhatsApp chatbot builder, or Vatio from your coding agent.
Vatio vs Intercom Fin: When a support team's helpdesk agent fits, when an agent built into your app does, and how to move.
Vatio vs Chatbase: A no-code agent trained on your content, or one your coding agent builds into your app, and how to move.
Vatio vs Voiceflow: A platform for chat and voice agents, or an agent defined in your app's project, and how to move.
Vatio vs Botpress: A bot platform with a studio and a TypeScript ADK, or an agent in your app's project, and how to move.
Vatio vs Tidio: A no-code support platform, or an agent your coding agent builds from your app, and how to move.
Vatio vs Manychat: Chat marketing on Instagram and other apps, or an agent for your app's signed-in users, and how to move.
Vatio vs the Vercel AI SDK: What you build and run yourself around the AI SDK's loop, what Vatio runs instead, and how to move your tools.
Vatio vs the Claude Agent SDK: An agent harness to host yourself, or a runtime for your app's customer chat, and how to move.

#Topics that live in the documentation

These have one page, in the reference, and no separate guide:

Human in the loop: when the agent hands a conversation to a person, and how it comes back.
Safeguards: the checks every reply passes before it is sent, and the rules you add.
Knowledge bases: what the agent answers from.
Your daily improvement routine: reports in the inbox, /vatio-improve, tests, publication.
Authentication and Tools: connect your backend and recognize signed-in users.

Latest CLI version: 0.114.0 · Changelog · The whole contract as Markdown: /docs/docs.md · or vatio docs