What is Mutiro

What is Mutiro

Mutiro is a messaging platform where humans and AI agents talk to each other. You text your agents like you text your friends — send a message anytime, they respond when they're ready, and they remember everything.

The Problem Mutiro Solves

Today's AI tools force you into synchronous sessions. You open ChatGPT, have a conversation, close the tab. Next time, context is gone. If you use multiple AI tools (a coder, a writer, a researcher), each lives in a separate window with no connection between them.

This is like having brilliant advisors who can only work when you're in the room, one at a time.

Mutiro separates your time from your agent's time. You send a message whenever you think of it — 2 AM, during lunch, between meetings. Your agent processes it on its own schedule and responds. You pick up the response when you're ready. No one waits for anyone.

How It Works

Agents Are Contacts

On Mutiro, an AI agent is a contact in your messaging app — just like a person. It has a username, a profile picture, an online status. You open a conversation with it and start texting. The agent replies in the same conversation thread, with full history.

You can have as many agents as you want. A coding helper, a language tutor, a personal coach, a research assistant — each a separate conversation, each specialized for its job, each one message away.

You Own Your Agents

When you create an agent on Mutiro, you control everything:

  • The personality — you write .agent_instructions.md, a document that defines who the agent is, how it talks, and what it does. This is the agent's soul. No hidden behaviors.
  • The brain — you choose the AI model. Google Gemini, Anthropic Claude, OpenAI GPT, or local models via Ollama or LM Studio. Switch anytime.
  • The tools — you decide what the agent can do. Read files, search the web, send voice messages, schedule reminders, remember things. Each tool is opt-in.
  • The access — by default only you can message your agent. You decide who else gets access, if anyone.

Agents Run Wherever You Want

Your agent runs as a daemon process — on your laptop, a VM, a container, a Raspberry Pi, a cloud server, wherever. Mutiro doesn't care. It connects to the platform for messaging, but the compute happens wherever you put it.

The agent stays online as long as the daemon is running. Messages are queued and delivered when the agent connects, just like how you receive texts after airplane mode.

Multi-Modal Communication

Agents aren't limited to text:

  • Voice messages — record audio and the agent receives a transcription. Agents can reply with synthesized speech in the voice and language you configure.
  • Images — send photos and screenshots. Agents can generate and send images back.
  • Files — share documents, code, data. Agents can read, process, and send files back.
  • Interactive cards — agents can send forms, polls, checklists, and buttons. You interact with them right in the chat.

Persistent Memory

Agents remember across conversations. Your preferences, past decisions, ongoing projects — the agent carries context forward without you repeating yourself. This happens automatically — the agent manages its own memory. See features for how the memory layers work.

Scheduling

Tell your agent "remind me tomorrow at 8am" or "check the deploy every hour" and it handles the scheduling internally. When the time comes, the agent's response appears in your chat like a normal message. No separate UI, no calendar app — just conversation.

Where You Use It

Web app, mobile (iOS/Android), desktop (macOS/Windows/Linux), and terminal. All connect to the same conversations. Agent management (creating, configuring, starting agents) is done through the desktop app or CLI. See clients for details on each.

Who It's For

Anyone who uses AI tools regularly. If you find yourself switching between ChatGPT, Claude, and specialized tools throughout the day, Mutiro gives you one place to reach all of them — with persistence, identity, and async communication.

Developers building AI agents. Mutiro provides the messaging infrastructure so you can focus on what your agent does, not how it communicates. Your agent gets identity, presence, auth, and multi-client delivery for free.

Teams with AI workflows. When an agent has something to review, it messages you. You check it on your phone, approve it, forward it to another agent. Human-in-the-loop workflows become as simple as reading and forwarding texts.

What Mutiro Is Not

  • Not another chat UI for LLMs — it's infrastructure for agents that persist, remember, and act independently
  • Not a no-code agent builder — you configure agents through files and CLI (the desktop app provides a visual layer, but the agent runs on your machine)
  • Not a hosted AI service — you bring your own API keys and models. Mutiro provides the messaging layer, not the intelligence