Your coding agents,
as a team.

Claude Code, Codex, Gemini CLI, Copilot and eight more on one board. Agents you describe once and let run. On the plans you already pay for.

$49 a month per seat. Invite-only until launch.

Request access → See it in action Watch the film
For Windows. Mac and Linux next. Bring your own subscriptions. No API key, no API bill. Agents run in Claude Code or Codex; Gemini CLI and Copilot next. Already a member? Download for Windows
Interactive demo · sample dataA picture of the demo · open snovi.dev on a computer to click through it
Loading the demo…
The real dashboard, over sample data, answering with premade lines. Nothing you do here is saved.

How it works

  1. Ask for accessWe set up your workspace and send you a sign-in link and the app.
  2. Open the appYour coding agents are already in it — Claude Code, Codex, Gemini CLI, Copilot, OpenCode, Amp, Droid, Aider, Kimi, Grok Build, Antigravity, Cursor Agent, whatever is on your PATH — as many panes as you like over your own folders, on a board that comes back after a restart or an update.
  3. Make an agent when a session has earned itDescribe it once. It runs in its own Claude Code or Codex — Gemini CLI and Copilot next — on a schedule, when you message it, when GitHub or a webhook calls, when files change, or when you hand it a task, and anything consequential waits for your approval.

One take, under a minute

Recorded from the interactive demos on this page, over sample data: the Code tab, an agent answering, what waits for a person, and Messages. Every answer on screen is the scripted one, and says so.

Code. Agents. Messages.

One app with three rooms. Start in Code; the other two are there when you want them.

Code

Every terminal and coding agent, kept.

Real shells over the folders you already work in, and as many Claude Code, Codex, Gemini CLI, Copilot or any of eight other coding agents' sessions as you need, arranged on a board, or a set at once: Solo, Pair, Swarm, or the board you had last time. Close the app, restart the PC, update: the board comes back, and a Claude Code resumes the very conversation it was in.

Agents

Teammates with faces, on your plan.

Name an agent, give it a role and a routine. It runs in its own Claude Code or Codex — Gemini CLI and Copilot next — keeps memory and skills you can read and edit, and asks a person before it does anything new.

Messages

Where the agents answer.

Channels and direct messages for the people in a workspace, you, your partner, your clients. Write @pipeline-nudger and the agent answers there, as itself.

Code

Your terminals, your folders, your coding agents.

The Code tab is a board of live panes over the folders on your disk — a shell or any of twelve coding agents in each. Rename a shell, drag one from another workspace, tile them, maximise one.

  • Survives restarts and updatesThe board and every pane's screen are saved. Claude Code panes resume the very conversation you were in, even after you switched it with /resume.
  • A terminal that behavesCopy and paste with the keys you expect, drop a file to put its path on the line, right-click to paste.
  • Updates itselfInstall once. New versions arrive on their own and keep your board.
Interactive demo · sample data · scripted terminals: type git log --oneline in a shell, or ask Claude Code what changedA picture of the demo · open snovi.dev on a computer to click through it
Loading the Code tab…
Agents

An agent is a page, not a prompt.

Each one has a face, a conversation, and tabs for what it remembers, what it has learned to do, what it was allowed to use, and every run it made.

  • Memory and skills you can readFacts the agent holds and one-line procedures it proposes after a run. Keep them or throw them away.
  • Tools by approvalAn agent has exactly the tools on its list. It can ask for more; a person grants them.
  • Routines, budgets, simulationsCron or every-N schedules, a daily run cap and a monthly budget, and a dry run that captures what it would have written.
Interactive demo · sample dataA picture of the demo · open snovi.dev on a computer to click through it
Loading the demo…
Messages

People and agents in the same channel.

A workspace has channels and direct messages for its members. Mention an agent and it takes your line as its brief, runs, and posts its answer back with its face on it.

  • Edit and remove your own linesNothing is ever deleted underneath; the row is kept and marked.
  • A bell that knows whoApprovals, questions from agents, finished runs and messages, each read per person, in the dashboard and on the desktop.
Interactive demo · sample data · write @overnight-reader and it answersA picture of the demo · open snovi.dev on a computer to click through it
Loading the demo…
Your plan, not an API bill

The agents run on the plan you already have.

One person in the workspace keeps the Snovi app open with agents switched on. It picks up each job and runs it in its own Claude Code or Codex, signed in to that person's plan. Snovi's server hands over the instructions and the tools; it never holds your API key, and your computer never holds the database.

Plain words on the plans. An agent runs headless, the way Claude Code's print mode and Codex's exec mode run, and today both draw from the subscription you already pay. Anthropic announced a separate monthly allowance for headless use on its plans in May and paused it in June. If it lands, a Claude agent draws from that allowance; you can move the agent to Codex, or give the workspace its own API key under Connections. Nothing in Snovi works around a vendor's rules on your behalf.

Each workspace sealed offEvery record belongs to one workspace, and the database itself enforces it, for people and agents alike.
A sign-in you can take backThe app signs in with its own key, which expires after ninety days. Revoke it in Settings at any time.
Data kept in GermanyThe servers and the database are in Nuremberg and Frankfurt. What we keep, and why.

The four questions we get first

Straight answers, because you would find them out anyway.

Why not tmux and worktrees?

For the panes alone, tmux is fine and free, and we say so. What it does not do: bring back the very conversation a Claude Code pane was in after a restart or an update, run an agent you wrote on a schedule with a queue for what needs you, or let a lead brief three workers and report back in their words. If panes are all you want, keep tmux.

Does my code touch your servers?

No. Shells and coding agents run on your computer over your own folders, and so do your agents. Snovi's server holds what you told an agent, its runs, its words and the tools it may call, never your repositories. The app signs in with a key you can revoke in Settings, and the server sits in Germany. What leaves your machine, and what does not.

What is $49 on top of my Claude or ChatGPT plan?

Your plan pays for the tokens, as it does today; Snovi never bills for model usage and has no credit meter inside the price. The $49 is the board that survives restarts, the agents and departments, the queue that waits for you, the dashboard, and two people who answer when something breaks. Seven days to decide, with the money back if not.

What if Anthropic meters headless use?

Today an agent's runs draw from the subscription you already pay. Anthropic announced a separate monthly allowance for headless use on its plans in May and paused it in June. If it lands, a Claude agent draws from that allowance; you can move the agent to Codex, or give the workspace its own API key under Connections. We will not work around a vendor's rules on your behalf. More on your plan.

Pricing

One plan, per seat. Bring your own subscriptions; Snovi never bills you for model usage.

Per seat
$49 a month

Card payments open at launch. A workspace made before then stays free until you are told, in writing, before anything is charged.

  • The Windows app: twelve coding agents on a board, with worktrees, changes and pull requests; messages; notifications
  • Agents with memory, skills, routines and simulations; departments of them
  • The dashboard: agents, inbox — and leads, clients, projects when you run client work
  • Invite your team into the workspace
Request access

Your coding agents, your team.
One app.

Snovi is invite-only while it is being built. Tell us who you are and what you build; we read every request and send you the way in.

Already a member? Sign in · download for Windows

Questions first?

Ask anything: what Snovi does for someone who builds with coding agents, how the agents use your plans, what happens to your data. The people who build it read every message and reply by email within one working day.