OmO
70.9kInstall

Your tool for real work.
But it's an agent.

Turn your tokens into results that matter: research across ten thousand sources, a deck people actually get, backends, frontends and code.

Get startedGet the desktop beta

From Sisyphus Labs, the people who made Dori and doriterm.

OmO — market research, 12k sources

$ mass ulw 12k sources → ranking model → deck

Workflow · market research, 12,000 sources

Wave 1 · 0/1 done
Wave 2 · 0/2 done
Wave 3 · 0/3 done
Wave 4 · 0/3 done
Wave 5 · 0/1 done

01Ultrawork

Add one word to a hard task. The agent takes it from there.

Skip the study. If a task feels hard, add the word ultrawork to your prompt. The agent decides how to attack it and drives to the best result on its own.

omo · ~/reports

› make the quarterly report actually readable ultrawork

  1. reads the whole codebase before touching a line
  2. plans, then proves each step before moving on
  3. verifies the result on the real surface, then stops

02Mass ulw

One keyword, hundreds of agents. Each part on the model that fits it.

Type mass ulw and the whole job becomes a graph of agents: deep research over thousands of sources, a full machine-learning pipeline, a migration nobody wants to do. Each part runs on the model that fits it, in parallel, and gets checked before anything is called done.

OmO — market research, 12k sourcesWorkflow

market research, 12,000 sources

wave 1/5 · 0/10 done · 7 models

queued
Wave 1 · 0/1 done
Wave 2 · 0/2 done
Wave 3 · 0/3 done
Wave 4 · 0/3 done
Wave 5 · 0/1 done

03Models

Frontier models and cheap ones, together. Each ships with a prompt tuned for it.

Use the frontier models you love together with the cheap ones. Just sign in. Every model ships with a system prompt tuned for it, so you never pick the wrong one.

  • Claude Fable 5.1 Max
  • GPT 6 Astra High
  • Kimi For Coding Highspeed
  • Grok 4.6 xHigh
  • GLM 5.2
  • DeepSeek V4.1 Flash
  • Claude Opus 5 High
  • GPT 5.6 Sol Medium
  • GPT 5.6 Luna Fast
  • Claude Sonnet 5
  • GPT 6 Astra Max
  • Claude Fable 5.1 Medium

Tuned profiles shipped with OmO

04Kibitzer

Your agent remembers your work. A small model keeps the memory.

A small, cheap model sits beside the expensive main agent, reads what you've done before, and nudges the main agent before it repeats a mistake. The expensive model stays sharp; you stop re-explaining yourself.

Kibitzer · small model

Small, cheap model · independent loop

  1. Watch
  2. Read memory
  3. Judge

memory

Previous migration: production broke when the test was skipped.

Main agent · frontier model

Expensive frontier model · main loop

  1. Plan the schema change
  2. Act · write the migration
  3. nudgeLast time this broke production. Run the migration test first.
  4. Verify · run the migration test first
  5. Ship the change

05Skills

Real-world know-how, as skills. Loaded only when the task needs them.

Your model is smart, but it still doesn't know a lot about solving real-world problems. OmO ships the missing knowledge as skills and loads only what the task needs.

  • frontenddesign system, layout, motion, visual QA
  • debugginghypothesis loop, failing test first
  • ulw-researchthousands of sources, claim graph, cited report
  • data-scientistDuckDB, Polars, charts from raw files
  • git-masteratomic commits, rebase, bisect
  • ultimate-browsingJS pages, logins, screenshots
  • browsera real browser, driven from code
  • visual-qaHIG checklist, light and dark, phone and desktop
  • ast-grepstructural search and codemods
  • review-workone reviewer, real QA evidence
  • refactorthe right decomposition for a cleanup
  • hyperplanadversarial planning before the work
  • visualizean answer as a self-contained HTML page
  • lsp-setuplanguage servers wired for the agent

See every skill in the docs

06Crafted

Everything else is crafted. You won't see most of it.

That's the point. Each of these runs under every session.

  1. 01
    Code-mode tool calls

    Independent tool calls run together in one small program instead of one slow round trip each.

  2. 02
    Tool-call absorption

    Dozens of read, search and lookup calls collapse into a single step. Fewer turns, less waiting, fewer tokens.

  3. 03
    Tool-call correction

    Malformed arguments are repaired before they execute, so a typo never becomes a wasted turn.

  4. 04
    Team mode

    A lead agent runs a team of specialists in parallel when a job is big enough to split.

  5. 05
    Monitors, not polling

    Waiting on a build, a deploy or a file is a subscription. The agent keeps working and wakes when it lands.

  6. 06
    Goals that survive

    Long work is checkpointed with evidence, so a session restart resumes where it stopped.

  7. 07
    Hot-reloading everything

    Config, skills and prompts reload live. No restarts to try a change.

  8. 08
    Computer use

    The agent reads the screen and accessibility tree of native apps on macOS, Linux and Windows, then clicks and types in the background, so your focus and cursor stay put. A stop chord suspends it; only you resume.

Read the details in the docs

Where it runs

Your terminal today. Your chat apps next.

In your terminal

One command installs OmO. Then run omo in any project.

macOS, Linux, WSL and Windows

In a desktop app

The OmO desktop app for your Mac, Windows PC or Linux machine.

Beta

From your chat apps

Talk to your agent where you already are. It messages as you, not as a bot, on every platform you use.

  • Slack
  • Discord
  • Teams
  • Webex
  • Telegram
  • WhatsApp
  • LINE
  • WeChat
  • iMessage
  • Instagram
  • KakaoTalk
  • Channel Talk
  • Lark

Coming soon. Messaging platforms are being wired in one by one.

What people say.

It made me cancel my Cursor subscription. Unbelievable things are happening in the open source community.
Arthur Guiot
If Claude Code does in 7 days what a human does in 3 months, OmO does it in 1 hour.
B, Quant Researcher
You guys should pull this into core and recruit him. Seriously. It's really, really, really good.
Henning Kilset

Made by Sisyphus Labs. The people behind Dori and doriterm.

DoriAn engineer who lives in your chat. Tell him what to do in Slack, Discord or Telegram; he gives each job its own coding-agent session and reports back when it is merged.sisyphuslabs.ai →doritermEvery machine you own, linked through an end-to-end encrypted mesh, in one terminal. Start OmO on the mac mini at home and answer it from your phone.doriterm.com →

One command. Then get back to work.

Install OmO, open your project, describe the job. That's the whole setup.

macOS, Linux, WSL

$ curl -fsSL https://get.omo.dev/install.sh | bash

Windows PowerShell

PS> irm https://get.omo.dev/install.ps1 | iex

Windows CMD

> powershell -ExecutionPolicy Bypass -c "irm https://get.omo.dev/install.ps1 | iex"

Then open a new terminal and run omo.Run this on your Mac, Linux or Windows computer.

Read the docsRead the manifesto