Sutando
My AI Stand. Realtime by day, rewriting itself by night. Summon my AI superpower.
Last verified:
What is Sutando?
Sutando is an open-source personal AI agent that runs 24/7 on your Mac, named after Stands from JoJo's Bizarre Adventure — a personal spirit that fights on your behalf. It handles voice commands, phone calls, meetings, and screen control, working autonomously while you sleep. The AI can be summoned by phone call to control your computer remotely, share screens to Zoom, record narrated videos, join meetings, make phone calls, and even build itself by acquiring new skills and generating its own avatar.
Key features include realtime voice control, phone summon capability from any device, screen sharing and capture, browser automation, meeting participation with narration and slide advancement, autonomous task execution (like catching CVEs, opening PRs, setting up Discord bots), voice-editing text at the cursor position, cross-instance communication between multiple Sutando instances, self-diagnosis and self-improvement by reading its own logs, and the ability to talk to other AIs (like making restaurant reservations through AI-to-AI negotiation). It integrates with Slack, Discord, Zoom, Odoo CRM, GitHub, and other desktop applications.
Sutando is designed for developers, tech professionals, business owners, and anyone who wants hands-free computer operation. It's particularly useful for people who need to operate their Mac remotely while away from their desk, developers who want AI to ship PRs autonomously, sales professionals who need AI to demo software, security professionals who need alerts handled while away, and anyone who wants to run their business hands-free using voice commands.
Sutando pricing
Pricing model: Freemium
Sutando itself is open-source under MIT License and free to use from GitHub. Voice functionality uses Gemini voice API (same key as Basic tier). The Free tier covers normal use at approximately 15 requests per minute. Heavy use requires paid Gemini tier at approximately $0.30 per use. The private beta is currently free for those granted access on macOS. No traditional subscription model — you pay for API usage through Google Cloud.
Sutando pros
- Open-source under MIT License with active GitHub development
- Runs 24/7 autonomously on your Mac with your data and hardware
- Voice-controlled computer operation from any phone call
- Screen sharing and Zoom integration for remote work
- Can join meetings and present slides with live Q&A
- Autonomous task execution like opening PRs and fixing bugs
- Self-improvement capability by learning from mistakes overnight
- Cross-instance communication between multiple Sutando instances
- Works with Slack, Discord, GitHub, Odoo, and other apps
- Voice-edit text at cursor position for tweets, emails, code
- AI-to-AI negotiation for tasks like restaurant reservations
- Security alerts handled automatically while you're away
- Remote file retrieval from Mac while walking or away from desk
- Creates narrated video explainers from breaking news autonomously
- No keyboard or mouse needed — entirely voice-controlled
- Private beta with macOS support
- Self-diagnosis reads logs and surfaces friction points
Sutando cons
- Currently in private beta — access requires request
- macOS only — not available for Windows or Linux yet
- Requires Gemini voice API key for voice functionality
- Very new project (created March 28, 2026) — unproven long-term
- Heavy use requires paid Gemini tier at ~$0.30 per use
- Free tier limited to ~15 requests per minute
- Technical setup required — runs as Claude Code session
- Requires constant Mac running for 24/7 operation
- Privacy concerns with AI watching screen and making calls
Frequently asked questions about Sutando
What is Sutando?
Sutando is an open-source personal AI agent named after Stands from JoJo's Bizarre Adventure — a personal spirit that fights on your behalf. It runs 24/7 on your Mac and handles voice, calls, meetings, and screen control. You can summon it by phone call to operate your computer remotely, share screens, join meetings, make phone calls, and it even builds itself by acquiring new skills autonomously.
How do I get access to Sutando?
Sutando is currently in private beta for macOS. You need to request access through their website at sutando.ai, and they will contact you when you're granted access. The open-source code is available on GitHub at github.com/sonichi/sutando under MIT License.
What operating systems does Sutando support?
Sutando currently supports macOS only. It is in private beta for Mac users and is not yet available for Windows or Linux systems.
How does phone summon work?
You can call your computer from your phone and Sutando will answer inbound calls. For outbound calls, you just ask and it will book tables, dial into meetings, or make calls. Verified-caller tiering ensures strangers stay information-only. In demos, users have called from their phone while walking miles from their desk and Sutando retrieved files, shared screens to Zoom, and controlled their Mac entirely by voice.
Can Sutando attend meetings for me?
Yes. Sutando can join your meetings, share your screen to Zoom, narrate presentations, advance slides, and field live Q&A. It has presented live on stage at conferences like ICLR 2026, delivering real talks as an AI agent with auto handoff, live demos, and audience Q&A.
Does Sutando improve over time?
Yes, Sutando builds itself and improves autonomously. It reads its own logs to self-diagnose and surface friction, fixes bugs overnight while you sleep, lists bugs it fixed and lessons learned from mistakes, writes its own rules into files that the next session reads, and acquires new skills as it learns your style. It can even name itself after earning real capabilities.
What apps and services does Sutando integrate with?
Sutando works with Slack (assign tasks that ship PRs), Discord (set up bots, reply to threads), Zoom (screen sharing, video recording), GitHub (open PRs, clone repos), Odoo (CRM, Sales, Accounting ERP), your spreadsheets, email, and anything on your desktop. It watches your screen and can operate any application through browser automation and GUI control.
What is the cost to use Sutando?
Sutando is open-source under MIT License and free to get from GitHub. For voice functionality, you need a Gemini voice API key. The Free tier covers normal use at ~15 requests per minute. Heavy use requires paid Gemini tier at approximately $0.30 per use. There's no traditional subscription — you pay for API usage through Google Cloud.
Can multiple Sutando instances work together?
Yes. Sutando supports cross-instance communication where multiple instances across different owners can teach each other. They can cold-review each other's PRs, perform cross-owner healing (one user's fleet recovered via another user's Sutando reaching across), and spread capabilities across instances. The project has spread to 5 instances across 3 owners who teach each other.
What are some real-world use cases for Sutando?
Real use cases include: catching CVEs while away (called phone, summarized advisory, pushed fix, opened PR), assigning Slack tasks that ship PRs without logging into GitHub, autonomous B2B sales demos in Odoo CRM, retrieving 157MB files from Mac while walking, running business hands-free with lead research and customer email, voice-editing tweets at cursor, setting up Discord bots without reading docs, turning breaking news into video explainers, and AI-to-AI restaurant reservation negotiation.