Free to start · No credit card required

The operating system for AI agents

Monitor every run. Host agents without infrastructure. Get alerted when things break. Ship AI agents with the confidence of a production system.

Works with any language or runtime — Python, Node.js, Go, or plain curl

email-classifier · run_7f3alive
00agent.run.started
01llm.request
02tool.called
03tool.returned
04llm.response
05agent.run.completed
06
6
events
1
tool call
1.2s
duration
Replay

Proof, not promises

The agents that run our company

These aren't demos. They're the actual agents operating AgentOS — the company — on AgentOS — the product — right now. Our inbox, our support queue, our outreach, our social feed, our on-call briefings: agents do the work, humans approve what matters.

Email Triage

runs nightly

Reads every inbound email, classifies it — support, sales, billing, spam — and files it with the right label. Genuinely ambiguous mail stays in the inbox for a human.

Support Reply

runs nightly

Drafts grounded replies to support email and sends the routine ones. Complex or sensitive threads become drafts for human review instead of auto-sending.

Outbound Outreach

on demand

Sends personalised outreach from the prospect sheet, tracks what was sent, and surfaces replies for a human to handle.

Social Content

runs Mondays

Drafts a week of LinkedIn and X posts every Monday and emails them for review. Publishing is approval-gated — nothing goes live without a human clicking approve.

Competitor Intel

runs Mondays

Fetches competitor blogs and changelogs, summarises what each one shipped last week, and emails a structured intel brief.

Incident Monitor

runs mornings

Summarises overnight alerts and incidents into a single what-happened-what's-open brief for whoever is on call.

Want agents like these? Install ready-made agents from the marketplace — one click, then connect your tools.

How it works

Hire. Connect. Supervise.

Launching an agent takes minutes, not sprints — and you decide exactly how much rope it gets.

1

Hire an agent

Pick a template or describe the job

Install a ready-made agent from the marketplace — the same ones running our company — or write a prompt describing the job. No infrastructure, no deployment, no code.

2

Connect its tools

Gmail, Sheets, GitHub, Slack & more

OAuth in two clicks — email, docs, CRM, code, calendars, social. Then choose which tools are approval-gated: the agent can read freely, but sending, posting, or spending waits for you.

3

Supervise, don't babysit

Approvals inbox, live runs, alerts

Agents run on schedules or on demand. Risky actions pause in your approvals inbox until a human decides. Every run is recorded end-to-end, and alerts fire the moment something drifts.

Features

Real work needs real guardrails

Human approval gates

Gate any tool behind sign-off. The agent pauses mid-run, your team gets notified, and nothing sends, posts, or spends until a human approves.

Hosted agents

Define the job, tools, and model in the dashboard. AgentOS runs the whole loop — no infra, no cold starts, no ops. Invoke with one HTTP call or a schedule.

Agent marketplace

Install proven agents in one click — including the ones running our own company. Skip the blank page; start from something that already works.

Every run on the record

Every event, tool call, and LLM response captured with millisecond precision. When an agent did something, you can see exactly what and why.

Smart alerts

Conditions on error rate, duration, silence, or custom events. In-app, email, or webhook — you hear about problems before your customers do.

AI reports & intelligence

Built-in intelligence agents watch your agents — anomalies, regressions, prompt weaknesses, token waste — and daily AI summaries keep you ahead of it.

import

TypeScript & Python SDKs

Already have agents in code? Instrument them in three lines — startRun, emit, complete — and they show up in the same dashboard, approvals and all.

Ingest API & OpenTelemetry

POST events from any language, or point your OTel GenAI traces at our OTLP endpoint. If it runs anywhere, AgentOS can supervise it.

MCP server

Connect Claude, Cursor, or any MCP client to your workspace. Manage agents, approve runs, and read reports without leaving your editor.

For developers

Already built agents? Keep them where they are — add three SDK lines and they get the same run history, approvals, and alerts as hosted agents.

$ npm install @agentos-sdk/core$ pip install pyagentos-sdkRead the docs →

Pricing

Cheaper than the work it replaces

Start free, no credit card. Upgrade when your agents earn it.

Free
$0forever

Put your first agent to work.

  • 2 agents
  • 1,000 runs / month
  • In-app alerts
  • Community support
Get started
Most popular
Team
$20/ seat / month

For teams handing real work to agents.

  • Unlimited agents
  • Human approval gates on any tool
  • All integrations — Gmail, Sheets, GitHub & more
  • Teams & per-folder access control
  • By-employee activity view
  • Email support

or $200/seat/year billed annually — 2 months free

Start with Team
Enterprise
Custom

For companies running agents at scale.

  • Everything in Team
  • Self-hosted / on-prem option
  • SSO / SAML & SCIM
  • Audit logs & data retention
  • SLA + dedicated support
Contact us

FAQ

Common questions

Do I need to be a developer to use AgentOS?+

No. Install an agent from the marketplace, connect your tools with OAuth, and it starts working. Writing your own agent is describing the job in plain language — the prompt, the tools it may use, and which of those need your approval. Developers get SDKs and APIs on top, but they're optional.

Can an agent do something risky without my sign-off?+

Not if you gate it. Any tool — sending email, posting to social, writing to a spreadsheet — can be marked as requiring approval. A gated call pauses the run, lands in your approvals inbox, and nothing happens until a human approves or rejects it. We run our own social publishing this way.

What actually runs the agent — do I need servers?+

AgentOS hosts and runs your agents — the LLM calls, the tool executions, the schedules, the retries. There's nothing for you to deploy or keep alive. If you'd rather run agents in your own code, the SDK reports them into the same dashboard.

Do I need to rewrite agents I already have?+

No. Add the SDK (npm or pip) and wrap your code with startRun/complete — about 10 minutes. Or export OpenTelemetry traces to our OTLP endpoint with zero AgentOS-specific code. Your agent keeps running wherever it is.

Is my data private?+

Yes. Run data is scoped to your workspace and only accessible by workspace members. Credentials for connected tools are encrypted at rest, and secrets are readable only by workspace admins. We do not use your data to train models. An on-prem / self-hosted option exists for Enterprise.

Which models can agents use?+

Hosted agents run on Anthropic and OpenAI models out of the box, plus any OpenAI-compatible endpoint (including self-hosted models on Enterprise). Agents you run through the SDK can use anything — AgentOS supervises the run either way.

Your first agent could be
working by this afternoon.

Install one of the agents that runs our company, connect your tools, and watch it work — with you in the approval loop from day one.