
Agents that improve themselves
Define an agent, connect tools and MCP servers, and deploy. The harness runs outside the sandbox and continuously learns from every run, so your agents become more reliable over time.
Agents need their own machine
Celesto provides secure, isolated computers that AI agents can use to browse, code, and get real work done.
The microVM that boots every Celesto sandbox. Run it on your own infrastructure — no account, no lock-in.
Same CLI, same SDK, zero infrastructure. Sandboxes boot in seconds and you pay only while they run.
Core features

Define an agent, connect tools and MCP servers, and deploy. The harness runs outside the sandbox and continuously learns from every run, so your agents become more reliable over time.

Hardware-isolated microVMs boot in about 500ms with a shell, filesystem, and full browser. Pause, resume, and snapshots preserve state, while network controls keep your data secure.

Ship agents under your own product, then expose them through the channels your customers already use — MCP servers, Telegram, APIs, and more. Deploy once, reach every surface.
Production sandboxes with the primitives that make long-running agents practical — forking, streaming, and instant recovery out of the box.
Branch a running sandbox into parallel copies. Test variants, roll back, or fan out work — without reprovisioning.
Cold start to running in under 3 seconds. No waiting for provisioning or image pulls.
Suspend sandboxes when idle — resume in milliseconds. Pay only for active compute.
Give agents a full browser inside the sandbox. Navigate, click, fill forms, and extract data — headless or streamed live to your screen.
Attached volumes survive restarts and snapshots. Your agent's data is always where it left off.
Full programmatic control. Execute commands, stream output, and manage sandboxes from code or terminal.
Execute AI code and agents in disposable sandboxes—built for workflows, long-running agents, and computer-use agents.
$ pip install celesto $ celesto computer create --cpus 2 --memory 2048 Name: einstein Status: running $ celesto computer ssh einsteinroot@celesto:/# python3 -c "print('Hello from the cloud!')"Hello from the cloud! $ celesto computer run einstein "uname -a"Linux celesto 6.8.0 x86_64 GNU/Linux $ celesto computer delete einstein✓ Computer einstein deletedNo setup needed — sign up and start running
Celesto provides an open-source framework and CLI for lightweight, secure sandboxes to run AI-generated code.
Live sandbox apps
Start Express, FastAPI, webhooks, dashboards, or an agent API inside an isolated computer. Publish a port and Celesto gives you a public URL - no separate deploy service to wire up.
Run the app and publish the port from Celesto. No Vercel-style detour or proxy glue.
Share an Express app, FastAPI server, dashboard, webhook receiver, or agent backend.
The service is reachable, but the runtime stays inside the disposable sandbox.
A public web app, served straight from the sandbox.
$ npm run dev -- --port 8000
$ celesto computer port publish
$ npm run dev -- --port 8000
ready on http://0.0.0.0:8000
$ celesto computer port publish
Port published
A public web app, served straight from the sandbox.
Agent observability
Every agent execution is automatically traced. Debug complex workflows, inspect tool calls, and optimize performance without writing a single log line.
Built into Celesto: every agent call is traced automatically, no configuration needed.
Watch traces stream in real-time as your agents run in production.
{"action": "get_order", "id": "12345"}{
"id": "12345",
"status": "shipped",
"carrier": "FedEx",
"tracking": "7890123456"
}No hidden fees. No surprises. Cancel anytime.
Try a real cloud sandbox with a clear monthly limit.
No credit card required
What's included
Keep a lightweight sandbox available all month.
What's included
For developers building and testing production agents.
What's included
For teams running many agents in parallel.
What's included
For secure agent infrastructure at scale.
What's included