Skip to main content
This guide uses the raw HTTP API. It’s the same surface every Orgo SDK wraps. Drop curl, requests, or fetch against https://www.orgo.ai/api and you’re done. No SDK required.
Prefer a helper library? Skip to Use the SDK after you’ve seen the raw calls. The SDK is a thin wrapper around the exact same endpoints.

​
1. Get your API key

Create an account at orgo.ai/start, then copy your key from orgo.ai/workspaces. 
export ORGO_API_KEY=sk_live_...
Every request takes this as a bearer token: 
Authorization: Bearer $ORGO_API_KEY

​
2. Create a workspace

Workspaces group related computers. 
curl -X POST https://www.orgo.ai/api/workspaces \
  -H "Authorization: Bearer $ORGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"name": "quickstart"}'
Response: 
{
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "name": "quickstart",
  "status": "active",
  "created_at": "2026-04-22T10:00:00Z"
}
Save the id. You’ll pass it as workspace_id when creating a computer.

​
3. Create a computer

curl -X POST https://www.orgo.ai/api/computers \
  -H "Authorization: Bearer $ORGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "workspace_id": "550e8400-e29b-41d4-a716-446655440000",
    "name": "agent-1",
    "ram": 4,
    "cpu": 1
  }'
Response (abridged): 
{
  "id": "a3bb189e-8bf9-3888-9912-ace4e6543002",
  "name": "agent-1",
  "status": "running",
  "ram": 4,
  "cpu": 1,
  "resolution": "1280x720x24",
  "url": "https://orgo.ai/workspaces/550e8400-e29b-41d4-a716-446655440000/computers/a3bb189e-8bf9-3888-9912-ace4e6543002"
}
The computer boots in under 500ms and is ready to accept commands immediately.

​
4. Take a screenshot

curl https://www.orgo.ai/api/computers/$COMPUTER_ID/screenshot \
  -H "Authorization: Bearer $ORGO_API_KEY"

​
5. Click, type, and run shell commands

# Click at (100, 200)
curl -X POST https://www.orgo.ai/api/computers/$COMPUTER_ID/click \
  -H "Authorization: Bearer $ORGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"x": 100, "y": 200}'

# Type text
curl -X POST https://www.orgo.ai/api/computers/$COMPUTER_ID/type \
  -H "Authorization: Bearer $ORGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "Hello, world!"}'

# Press Enter
curl -X POST https://www.orgo.ai/api/computers/$COMPUTER_ID/key \
  -H "Authorization: Bearer $ORGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"key": "Enter"}'

# Run a shell command
curl -X POST https://www.orgo.ai/api/computers/$COMPUTER_ID/bash \
  -H "Authorization: Bearer $ORGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"command": "ls -la"}'
Full list of actions: Mouse · Keyboard · Scroll · Bash · Python exec · Wait · Drag.

​
6. Let an AI drive it

Orgo exposes an OpenAI-compatible endpoint at /api/v1/chat/completions. Point any OpenAI SDK at https://www.orgo.ai/api/v1, pass computer_id, and the model will screenshot, click, and type on its own until your instruction is done. 
curl https://www.orgo.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $ORGO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "computer_id": "a3bb189e-8bf9-3888-9912-ace4e6543002",
    "messages": [
      {"role": "user", "content": "Open Chrome and search for AI news"}
    ]
  }'
Stream the agent’s progress token-by-token with "stream": true. See Create chat completion for the full spec, including thread continuation, custom Anthropic key, and error handling.

​
7. Lifecycle

# Stop  (auto-stop also does this after idle timeout)
curl -X POST https://www.orgo.ai/api/computers/$COMPUTER_ID/stop \
  -H "Authorization: Bearer $ORGO_API_KEY"

# Start a stopped computer
curl -X POST https://www.orgo.ai/api/computers/$COMPUTER_ID/start \
  -H "Authorization: Bearer $ORGO_API_KEY"

# Restart (reboot)
curl -X POST https://www.orgo.ai/api/computers/$COMPUTER_ID/restart \
  -H "Authorization: Bearer $ORGO_API_KEY"

# Delete permanently
curl -X DELETE https://www.orgo.ai/api/computers/$COMPUTER_ID \
  -H "Authorization: Bearer $ORGO_API_KEY"
Auto-stop is disabled by default. Configure it per-computer with auto_stop_minutes. See Auto-stop.

​
Use the SDK

If you’d rather not wire HTTP calls by hand, the official SDKs wrap this exact surface. They give you typed helpers, connection pooling for VNC, and one-line computer.prompt("do X") for the AI agent loop. 
pip install orgo

from orgo import Computer

computer = Computer(workspace="quickstart")

# Each of these is a direct wrapper over a single HTTP call.
computer.left_click(100, 200)
computer.type("Hello, world!")
computer.key("Enter")
output = computer.bash("ls -la")

# One-liner for the AI agent loop
computer.prompt("Open Chrome and search for AI news")

computer.destroy()
Both SDKs read ORGO_API_KEY from the environment and default to the same base URL you’d call directly.

​
Next steps

API Reference

Every endpoint, every field.

Use any model

Claude, GPT, Gemini, Hermes. Any OpenAI-compatible model.

Embed computers

Drop a live VM into your own app via VNC.

Auto-stop

Suspend idle computers; resume on next call.