Is yottoCode made by Anthropic?

No. It's an independent product from MXBO STUDIO and isn't affiliated with or endorsed by Anthropic. It uses Claude Code as the underlying technology and needs your own Claude Code subscription.

How much does yottoCode cost?

There's a free tier with 1 bot and 100 messages per month, and a Pro tier with unlimited bots and messages starting at $1.99 per month. The agent features also require a separate Claude Code subscription. Budget for both.

Which platforms can run yottoCode?

It's macOS only, and specifically macOS 13.0 or later on Apple Silicon (M1 and up). There's no Windows, Linux or Intel Mac build.

Can I use several projects at once?

Yes. Each bot handles one project directory, and the bundled multi-bot launcher runs multiple bots at the same time so you can work across repos.

Does it need API keys?

No separate API keys for the agent. yottoCode works with your Claude Code subscription, and you only need the Telegram bot token you create through @BotFather.

Is my code uploaded anywhere?

The app runs locally on your machine and routes messages outbound through a relay with no exposed ports. Your project files stay on your disk rather than being moved into a hosted container. yottoCode is a narrow tool aimed at a specific setup: you own a Mac, you already pay for Claude Code, and you want to run that agent from Telegram instead of a terminal. If all three fit, it removes a real annoyance by putting approvals, model switching and cost tracking into a chat you already carry around all day. That's the whole pitch. If any of those don't match, the Apple Silicon requirement and the second subscription make it an easy pass until a broader build shows up.

yottoCode

MXBO STUDIO · Coding

yottoCode is a native macOS app that bridges Claude Code and Telegram, letting you drive an AI coding agent from your phone or any device with a Telegram account. Think of it as Claude Code remote control: the agent runs on your own machine rather than a sandbox, so it keeps your files, tools and terminal, and you approve every risky action through inline keyboard buttons. The app ties into the official Anthropic Agent SDK and works with a Claude Code subscription instead of separate API keys.

Visit Tool

About yottoCode

What Is yottoCode

yottoCode is a desktop bridge between Claude Code and Telegram, built on the official Anthropic Agent SDK. The idea is simple: Claude Code is powerful, but it lives in a terminal on your Mac. yottoCode turns it into a Telegram bot for Claude Code, so you can send a request from your phone and watch the agent work on your actual development environment.

It isn't a cloud coding service. Your project stays on your disk, and the bot talks to it through an outbound-only relay, which means no exposed ports and no upload of your codebase to someone else's container. The developer, MXBO STUDIO, states clearly that the app is independent and not affiliated with Anthropic.

The biggest limitation to plan around is the subscription stack. yottoCode itself has a free tier and a paid Pro tier, but the agent features need your own Claude Code subscription on top. There's also a hard platform line: it's macOS 13.0+ on Apple Silicon only, so Windows, Linux and Intel Macs are out for now. That narrows the audience a lot.

Getting Started

Create a bot in Telegram by opening @BotFather, then copy the token it hands back.
Launch the yottoCode app on your Mac and paste in that token.
Pick the project directory you want the agent to work inside and choose a model (Haiku, Sonnet or Opus).
Message your bot to start a session; Claude Code replies with full agent capabilities and asks for permission before it edits files or runs commands.
Reply to any message to add context, and switch models or resume sessions from the same chat whenever you need.

Product Information

A quick look at yottoCode's pricing, supported platforms, and performance.

Free PlanYes

Paid Plans$0 - $1.99/mo

PlatformmacOS 13.0+ (Apple Silicon M1 and later)

DeveloperMXBO STUDIO

CategoryCoding

Release DateJun 2025

Latest UpdatedSep 2025

Website Visits1K

Website Global Rank14M

API AvailabilityN/A

Best for

The users, tasks, and scenarios where this tool fits best.

Users

Solo developers on a Mac
Developers who step away from the desk
People already paying for Claude Code

Tasks

Approving edits and commands remotely
Switching models mid-conversation
Watching cost as you go

Scenarios

Kicking off a deploy before you leave the office
Debugging on the move
Running weekend project work from a tablet or phone

Key features

Two-Way Telegram Chat With Voice

yottoCode handles text and voice in both directions, so you can dictate a request and have the agent talk back. That's handy when you're away from a keyboard and want to fire off a quick instruction without typing it out.

Interactive Permission Keyboards

Every sensitive action, like editing a file or running a shell command, comes through as a Telegram message with buttons to approve or reject. That's the core of Claude Code permissions in yottoCode: you choose between interactive approval and a full auto-allow mode, and you can flip between the two with a single tap. No surprises.

Session Resume and Model Switching

Sessions keep their full history, so you can pick a conversation back up later without losing context. From the same chat you can switch between Haiku, Sonnet and Opus, which lets you match the model to the weight of the task instead of restarting the session.

Full Desktop Access, No Sandbox

Because the app runs on your machine, the agent reaches your real files, tools and terminal. It can install packages, run tests, deploy and debug against your actual project state, which is the piece cloud Claude Code and plain AI chatbots can't touch. Big difference.

MCP Servers, Plugins and Cost Tracking

yottoCode connects to MCP servers and plugins, extending what the agent can do beyond its defaults. For anyone who leans on custom tooling, that MCP server control is the reason to pick it over a plain chat bridge. Live cost tracking sits alongside, giving you a running view of what each session is burning.

Smart Context Through Replies

Replying to a specific message gives Claude that message as context, and the agent reacts with an emoji to acknowledge what you sent. Curious how that feels in practice? It reads like a normal Telegram conversation rather than a command line.

Multi-Bot Launcher

Each bot maps to one project directory, which keeps things tidy but limits a single bot to a single repo. The included multi-bot launcher runs several projects at once, so you're not stuck choosing which repo to work on.

Pros and cons

Pros

Runs on your own Mac, so the agent uses your real files, tools and terminal instead of a sandbox.
Interactive permission keyboards keep you in control of every file edit and command.
Works with a Claude Code subscription, so there's no extra API key to buy.
Session resume, model switching and live cost tracking are built in rather than bolted on.
Voice input and output make it usable when you don't have a keyboard handy.

Cons

macOS 13.0+ on Apple Silicon only. Windows, Linux and Intel Mac users can't run it at all.
A separate Claude Code subscription is required on top of yottoCode, which raises the real cost.
One bot maps to one project directory, so anything more requires the multi-bot launcher.
The free tier caps you at 1 bot and 100 messages per month. That runs out fast.

Frequently asked questions

It connects Claude Code to a Telegram bot so you can send tasks, approve actions and follow the agent's work from any device. The agent still runs on your Mac, not in the cloud.

Forefront

Forefront · Coding

Forefront is a web platform for building with open-source AI. It lets you fine-tune leading open-source language models on your own data, evaluate how they perform, and run them through an API or export them to host yourself. Developers who want the convenience of a closed-source platform but insist on owning their models and data are the target audience here.

Free / $0 - $99/moView details

Startkit

StartKit.AI · Coding

Startkit is a boilerplate for building AI SaaS and AI wrapper products. Think of it as an AI startup boilerplate with the boring parts already wired up: authentication, Stripe and Lemon Squeezy payments, usage limits, transactional email, and an AI API starter that talks to OpenAI, Anthropic, Groq, or Llama. You clone the repo, set your price, and start on the part of your product that people actually pay for. It's Next.js under React and Tailwind, so most of the boilerplate code already feels familiar.

Paid / $99 - $499 one-timeView details

Testim

Tricentis · Coding

Testim is an AI-powered test automation platform for building and running end-to-end tests across web, mobile, and Salesforce applications. It leans on machine learning to keep tests stable when an interface changes, so teams spend less time fixing broken selectors. Not bad for an automated testing tool you can start using today. You create tests by recording actions in a browser, then optionally add JavaScript when you need more control. It's a solid pick for busy QA teams.

Free / Custom pricing on requestView details