Quickstart
Set up your workspace and hire your first agent in a few minutes.
This guide takes you from zero to a working agent in a few minutes.
1. Get Padiso
- Desktop app, download Padiso for macOS, Windows, or Linux and run it on your own machine.
- Cloud, deploy to your AWS, Azure, or GCP account using a blueprint.
2. Create a workspace
Sign in and create a workspace. A workspace is your isolated environment, it holds your agents, projects, tasks, runs, connections, and spend. Everything an agent can see or do is scoped to the workspace.
3. Connect a tool
Add at least one integration so your agents can do real work. Connections use the Model Context Protocol (MCP), so you wire a tool once and every agent can use it.
- Browse the integrations directory for tools, models, and clouds.
- Expose your own internal APIs or databases as MCP tools.
4. Hire your first agent
Create an agent and give it:
- a role (for example, "Support triage"),
- a runtime (the model or engine it runs on),
- a goal and a few starter tasks,
- a monthly budget.
5. Let it run
The agent wakes on its heartbeat, reads its workload, takes the most useful next action, updates its tasks, and reports back. You watch outcomes on the dashboard instead of prompting.
Gate any sensitive action behind an approval so a human clears it before it happens.
Next steps
- Concepts, how agents, tasks, and runs fit together.
- Governance, budgets, approvals, and audit.
- Deployment, local vs cloud.