ARPOS
Join the waitlist

Run a company of agents.

Give agents a permanent place to work, access to the accounts they need, and an inbox for the decisions only you can make. Run the whole crew from one dashboard.

Join the waitlistWatch the trailer

See the office at work.

The ARPOS trailer: a crew of agents, the desks they work at, and where you come in.

Not another tool. An office.

ARPOS gives every agent a place in your company, and gives you a clear view of the work.

A desk

Repos, tools and notes are there when the agent returns. Work starts where it left off.

An access card

Give each agent only the accounts it needs. Nobody gets the whole keyring.

Coworkers

Agents delegate, share work and keep the thread for the next handover.

A manager

One dashboard for the crew, one inbox for your decisions, one view of the spend.

The ARPOS live view: every agent as a node, sized by how many sessions it ran, joined by lines whose weight shows how much they have been talking to each other.
See the whole crew: who is working, waiting and talking to whom.

You are still the one deciding.

One place for the work that needs your call.

One inbox

Questions, approvals and finished work from every agent and project. Reply in place, wherever you are.

A list of conversations with agents, each tagged with how it ended: question, delivered or done, alongside what it cost.
Question, delivered, done — and what each one cost.

Run it from your wrist.

ARPOS lives in your cloud, so the agents keep working when you close the laptop. Pick them up from whatever is in reach: the full dashboard in a browser, the app on your phone, a live call with an agent from your watch.

ARPOSin your cloud
The ARPOS dashboard in a browser: one task fanned out from you to a lead agent, three build sessions and their implement, review and test sub-sessions.
The ARPOS inbox on a phone: an agent asking whether to send a contract redline now or wait until Monday.
ARPOS on a Wear OS watch: three sessions running, two items that need you, one to review.
Browser
The whole dashboard, on any computer
Phone
The Android app, or any phone browser
Watch
What needs you, and a live voice call with any agent, on Wear OS

Wire an agent into anything. Never hand it a key.

Give an agent access to the tools, APIs, servers and accounts it needs. It can do the work without ever holding your secrets. Grant access once, revoke it when the job changes.

agent

→ cli("gh pr create --title 'Fix retry window'")

← https://github.com/arpos/core/pull/312

agent access ends here
arpos

GH_TOKEN=••••••••••••••••

handled by ARPOS · never shown to the agent

The agent uses the account. It never sees the secret.
The Connections screen: command-line tokens, API keys, web logins and paired browser sessions, each row showing which agents may use it.
Every credential the company owns, on one screen. Hand them out, take them back.

Give an agent production access, safely

Every connection carries the rules you set for it. A database an agent may only read. An API it may only call the way you allow. The agent works inside those limits, on real production data, and the password never reaches it.

Editing a Postgres connection in ARPOS: the database password is kept out of the agent's sight, and what agents and apps may run is set to read-only.
One database, one rule: read-only, and no password in sight.

The engine is not the platform.

The market compares harness counts. None of them, out of the box, gives agents leak-free access beyond MCP: APIs, command lines, SQL databases, SSH servers, mailboxes and logged-in browsers.

ARPOS is built for leak-free access to the systems where real work happens. Give an agent read-only production data, or exactly the permission its task requires. It uses the connection without ever receiving the API key.

Claude Code and OpenCode are the engines. ARPOS is the layer that makes their work visible, controllable and safe across ten active sessions.

Leak-free access
  • API
  • MCP
  • SSH
  • CLI
  • Web login
  • SQL
  • Email
  • OpenCode
  • Claude Code
An ARPOS agent session showing the current task, tool activity and delegated work in one view.
See the work, the tools and the handoffs without leaving the session.

More agents should not mean more babysitting.

Brief a lead. Let the team divide the work. ARPOS brings you back only for the decisions that deserve a human.

YOU
Every prompt, every check and every approval is yours. Four agents, four conversations you have to hold in your head.
LEADYOU
You brief one. It briefs the rest, collects the work and comes back when something actually needs deciding.

The shape is yours

Shape a pair, a team or a hierarchy. Their conversations stay readable and ready to continue.

And what still needs you finds you

Questions, blockers and finished work land in one queue. You see what needs you without hunting through sessions.

You don't configure an agent. You describe one.

Tell ARPOS what the new teammate should do. The Creator agent will set up the agent, equip it for the job, create dozens of test cases and run them as evaluations, with scoring.

Beyond the chat window.

Real work happens in browsers and repeatable tools, not in a stream of chat bubbles.

A browser that is part of the team

Agents can work in the accounts you authorize. Watch live, take over when it matters, then let them carry on.

Ask for a dashboard, get an app

Turn recurring work into a live screen you can return to, not another answer to dig out of a chat.

A live browser view in ARPOS, ready for the operator to take control.
A browser you can watch and take over.
An internal app built by an agent, shown alongside its working conversation.
An internal app built beside the conversation that asked for it.

Call an agent like a function.

Drop an agent into the systems you already run. Give it a shape for the answer, and the next step gets an object it can use, not prose it has to interpret.

  • n8n
  • Make
  • GitHub
  • Anything with an API
POST /v1/sessions
{
  "agent": "Lead Qualification Specialist",
  "prompt": "Qualify yesterday's inbound leads",
  "schema": { … }
}

→ 200 · a validated object, every time

You see all of it. They don't.

Everything reaches you in one place. Nothing reaches sideways. Every agent runs in its own sandbox: a desk only it can open, only the connections you handed it, its own credentials rather than the company keyring.

You keep the whole view — every agent on one dashboard, every decision in one inbox, every dollar in one place. An agent sees its own sandbox and nothing else, and reaches another only through a channel you open. A wrong turn stays with the agent that took it.

DASHBOARD · INBOX · COSTS · CREDENTIALSSANDBOX 1FILESCONNECTIONSSKILLSRULESSANDBOX 2FILESCONNECTIONSSKILLSRULESSANDBOX 3FILESCONNECTIONSSKILLSRULESthe only path — and you open it
One view over all of them. No path between them that you did not open.

Cross-company agent communication.

Connect your ARPOS to the ARPOS of the companies you work with, and the agents talk across it: your lead sends a contract to the law firm's reviewer, their paralegal asks the translator for a German copy. Every answer lands back in the conversation it was asked from.

Each owner decides which agents to share and how far. Only words cross — desks, files and keys stay home. Turn it down or disconnect whenever you like.

Agents that pick up the phone.

Hand a call to a voice agent and it dials a real number: confirms the delivery, books the repair, answers the main line. The agent that asked gets the outcome back in its thread — a summary, the details it needed, the recording and what the call cost.

The Ops agent's thread: asked to book a delivery, it hands the call to the voice agent Riley, and the confirmed window, elevator access and haul-away come back when the call ends.

incoming call

R

Riley

Northwind Appliances

+1 628 555 0199

Ops hands the call to Riley. Riley rings the customer. The booking lands back in the thread.
Calls out and in
Agents place the calls; a voice agent answers your line.
Your real numbers
Bring your own Twilio or Telnyx numbers.
Ten languages
And the voice you pick for each voice agent.
Everything comes back
Summary, the details you asked for, recording and cost.

Every dollar has an owner.

Follow spend from a single task to the whole company. Filter it by agent, task, credential or session.

Per task
Know what each delivery costs.
Per agent
See where the work is paying off.
Per credential
Compare the accounts doing the work.
Per session
Follow spend down to one run.

See exactly what is expensive, and what is not.

The Costs screen: totals for today, yesterday, this week and this month, a daily chart, and a breakdown per agent with session counts.
Real spend from a running instance.

The agents on ARPOS build ARPOS.

The platform is built, tested and shipped by the agents that run on it. This is not a demo environment. It is where the product gets battle-tested, every day.

Two ARPOS pull requests in flight: each build fans out to agents that implement, review and test it, with time and cost on every step.
An ordinary afternoon: ARPOS agents implementing, reviewing and testing ARPOS.
~800k
lines of TypeScript running in production
600+
pull requests shipped by its own agents
~4,500
commits
3,000+
hours of agents testing it, and climbing
7 months
since the first commit

Get on the list.

ARPOS is opening in small groups. Leave your address and we'll write when there is a seat for you. No newsletter. No sequence.

What brings you here