Guide
Build your first routine.
Create a repeatable agent task and understand schedule and webhook behavior.
Start with a successful manual run
Write a narrow task that produces a file or message you can inspect. For example: summarize new notes from one folder into a weekly brief and list anything uncertain. State the input period, output location and actions that need approval.
Add a schedule
Choose the timing and check the next due run in the app. Review the first scheduled result. A schedule is a trigger, not proof of completion; inspect the run record and configure the surrounding workflow to notice missing results.
Trigger from another service
Use the routine’s webhook URL when an external event should start work. Treat the URL as a capability and do not post it publicly. Configure the sender to retry temporary failures, including a 503 while the hosted sandbox wakes. Deliveries are not queued during waking.
Keep exploring
A practical guide to working with agents.
Learn the basics of AI agents, set up Suparagent and build reviewable workflows.
ProductGive every agent a place on the team.
Explore named agents, teams, memory, routines, skills and review controls in Suparagent.
Security & dataUnderstand where the work runs.
Learn the boundaries between your browser, the hosted service, sandbox providers and AI engines.
Give the work a home
Build your first AI team.
Start with one agent, one clear task, and a result you can review.
Preview · Bring your sandbox provider and AI account