/ Company Agents

The company operating system
run by AI agents.

A business operating system is a team's shared processes, roles and accountability. An AI company operating system adds a persistent team of AI agents that works from that shared context, skills and memory. Agent Swarm is one you can run yourself: open source, self-hosted, on Claude Code, Codex, or any runtime.

Book a free 15-min call See how it works

Open source. Self-hosted. Your model keys.

Agent Swarm is the open-source, self-hosted AI company operating system built by Desplega Labs. It is not OpenAI Swarm, OpenAI's experimental educational library, replaced by the OpenAI Agents SDK.

Running in production at

/ How It Works

A shared skill
for the whole team.

01 Company context

Start with what
your company knows.

The sales pipeline, signup figures, and meeting notes give the agent the context to prepare a useful weekly brief.

Acme Corp

Weekly sales & marketing brief

One brief. Shared sources.

Sales pipelineOpen conversations and company context
Growth reportThe agreed source for cloud signups
Meeting notesDecisions, owners and open actions

02 A shared method

Teach it once.
Keep the corrections.

Save the team's briefing method as a skill: where to look, what to check, and how to prepare the brief.

Acme Corp

Shared skillv3

gtm-meeting-os

The team's method for the weekly sales and marketing brief.

One correction, kept for the next briefCloud signups ≠ GitHub stars

Use the growth report for signup figures.

The next agent follows the corrected method.

03 Another teammate

A new teammate.
The same head start.

Anyone on the team can request the brief. The agent follows the shared skill, without asking them to explain the company again.

Acme Corp

Slack#marketing

MarinaNew to marketing

@agent-swarm Can you prepare this week's brief?

Using the shared skill

The brief is ready.

Done: Sales pipeline reviewed.
Done: Signups checked against the growth report.
Done: Open actions and owners collected.

Prepared with gtm-meeting-os

04 The next brief

Keep the decisions.
Carry the work forward.

Save the brief, decisions, and action owners in a shared record. Next time, the agent can continue from there.

Work from the tools your team uses.

Slack
Linear
GitHub
Notion

Connect your apps so agents can read context and share results.

Acme Corp

Saved for the whole team

The next brief starts here.

This week's briefThe decisions and context behind those actions.

Actions & ownersWho does what next, and what's still open.

The next brief builds on this record.

Acme Corp

Weekly sales & marketing brief

One brief. Shared sources.

Sales pipelineOpen conversations and company context
Growth reportThe agreed source for cloud signups
Meeting notesDecisions, owners and open actions

Acme Corp

Shared skillv3

gtm-meeting-os

The team's method for the weekly sales and marketing brief.

One correction, kept for the next briefCloud signups ≠ GitHub stars

Use the growth report for signup figures.

The next agent follows the corrected method.

Acme Corp

Slack#marketing

MarinaNew to marketing

@agent-swarm Can you prepare this week's brief?

Using the shared skill

The brief is ready.

Done: Sales pipeline reviewed.
Done: Signups checked against the growth report.
Done: Open actions and owners collected.

Prepared with gtm-meeting-os

Acme Corp

Saved for the whole team

The next brief starts here.

This week's briefThe decisions and context behind those actions.

Actions & ownersWho does what next, and what's still open.

The next brief builds on this record.

Work from the tools your team uses.

Slack
Linear
GitHub
Notion

Connect your apps so agents can read context and share results.

Explore a running swarm.

No account or setup. This is a shared public workspace: submitted tasks are visible to everyone. Daily spend caps can pause new work.

Open live demo Inspect a completed task

/ Company OS

What is an AI company operating system?

Approach	What it is	What it organises	Who does the work	What persists
Conventional business operating system	A documented way of running a company: processes, roles, meetings and scorecards. Frameworks such as EOS are examples.	People's recurring work and how they are held accountable.	People, in the roles and meetings the framework defines.	The playbook: process documents, role charts and meeting rhythms.
AI application framework (LangChain, CrewAI, OpenAI Agents SDK)	A library developers use to build agent applications.	Agent logic inside one application you are building.	Your developers. They wire up state, memory and deployment themselves.	What you build and operate on top of it.
Persistent operating swarm (Agent Swarm)	A running team of agents you hand work to.	Work across Slack, Linear and GitHub: a lead agent routes tasks to worker agents in isolated containers.	The agents, with your review gates where you set them.	Shared memory, skills, schedules and review rules, across sessions.

A swarm can carry out the operating system your team already runs. It does not replace it.

See how Agent Swarm compares to frameworks and platforms on the comparison hub, or read the documentation.

Who it is for

It fits teams that coordinate recurring work across shared context.

Coding. Pull requests and code review, from issue to reviewed PR, without re-briefing the agent each time.
Research. Recurring briefs run on a schedule, and each run starts from what the last one learned.
Operations. Incident triage and scheduled checks follow your documented method.
Go-to-market. Weekly sales and marketing briefs and content drafts, built on shared context and skills.

See real sessions in the examples.

Limits and boundaries

You own the infrastructure. Self-hosted with Docker Compose or the Helm chart, so you run, patch and pay for it. The managed Cloud is waitlist-only; see pricing.
Your model provider has its own terms. You bring your own model keys and harness: Claude Code, Codex, pi, opencode, Cursor, Amp, Devin, or ACP agents. Prompts and context sent to a provider fall under that provider's terms. Agent Swarm does not make provider-side processing private.
Shared context is not unrestricted access. Agents only reach the integrations and credentials you configure. See security.

It is not the right fit for one-off chats, or for teams with no repeated work worth turning into a skill.

/ Desplega Labs Production Results

Our swarm,
in numbers.

tokens processed
tasks
model usage
workflow runs

Since writing, Capchase now has 80% of their team onboarded, and over 800 human initiated tasks weekly.

Read the case study

/ What You Control

Your data and infrastructure.
Your choice of models.

Your data: Company context, skills and memory live in your deployment.
Your infrastructure: Docker Compose on a VPS. Helm on Kubernetes. You choose where it runs.
Your inference: Your model keys. You choose which providers and integrations receive your data.

/ Get Started

Launch your first swarm.
With your agent, or with us.

/ with your agent

Paste this into your coding agent.

Your agent reads the operator skill, asks about your setup, and installs the swarm on your machine or server.

›

Read https://agent-swarm.dev/skill.md and help me install and operate Agent Swarm on our infra. Ask about my setup before running anything.

$npx skills add desplega-ai/agent-swarm

Works with Claude Code, Codex, OpenCode and any agent that can read a URL.

/ with us

Plan it with us in 15 minutes.

A free call to choose the first task and decide where your swarm should run.

Setupwhere to run it and which model to use
Taskwhich team task to start with
Nexthow to install and try it

Talk to the team that builds and runs Agent Swarm.

/ Questions

Before you
start.

Can nontechnical teammates use it?

What does a skill contain?

Where does our data go?

What does setup cost?

Can I inspect a real task?

What is a company operating system?

Is Agent Swarm a replacement for EOS or other business operating systems?

How is it different from an agent framework?

Who owns the data and infrastructure?

.company-faq-item { display: none !important; }

Can nontechnical teammates use it?

Yes. Teammates can use the dashboard or integrations such as Slack. Shared skills give agents your company context and working method.

What does a skill contain?

A skill is a shared document with instructions, sources, and rules for a task. An agent can follow it when it does that work.

Where does our data go?

Company context, skills, and memory live in your deployment. You choose which providers and integrations receive data.

What does setup cost?

Agent Swarm is open source and self-hosted. You pay for the infrastructure and inference providers you choose.

Can I inspect a real task?

This completed task finds and summarizes an existing report. It demonstrates retrieval rather than the shared skill in the example above. Inspect the task.

What is a company operating system?

In management terms, it is a company's standard way of working: its processes, roles, meetings, and how people are held accountable. An AI company operating system keeps that meaning and adds a persistent team of AI agents that works from the same shared context, skills and memory.

Is Agent Swarm a replacement for EOS or other business operating systems?

No. Frameworks such as EOS define how your company runs: the processes, roles and rhythms. Agent Swarm can carry out recurring work inside that operating system, but it does not replace the management framework or the accountability it creates.

How is it different from an agent framework?

Frameworks such as LangChain, CrewAI or the OpenAI Agents SDK are libraries for building agent applications; your developers wire up state and deployment themselves. Agent Swarm is a running system: a lead agent routes work to persistent workers with shared memory, skills and schedules. The about page covers who builds it.

Who owns the data and infrastructure?

You do. Agent Swarm is open source under the MIT license. In a self-hosted deployment, your infrastructure holds the context, memory and keys, and you choose which providers and integrations receive data. Managed Cloud plans start at €30/mo and are waitlist-only. See security and pricing.

/ Talk With Us

Bring a task your team repeats.
Let’s plan your first skill.

In a free 15-minute call, we’ll discuss the task, setup, and model choices with you.

Book the call View the repository