Routines
Routines run an agent for you on a schedule. You choose how often it runs, which tools it may use each time, and how deep it should think. You can then check the history when you want to see what it produced.
You create and edit routines from the same chat workspace you already use, so it should feel familiar if you chat with agents today. Open Routines from the chat sidebar to manage schedules and run history in one place.
📝 Note: Which schedules, tools, and plans are available depend on your organization and subscription. If you do not see routines, ask your organization admin.
What you can do
- Pick a schedule — Daily, weekly, monthly, or a custom pattern (for example every Monday morning or the first day of the month).
- Limit tools per run — Give the routine only the tools it needs so each run stays predictable and lower risk than turning on every tool at once.
- Adjust reasoning effort — Trade speed and cost against depth for automated runs versus when you are chatting live.
- Pick an approved model — When you choose a model directly on a routine, the list reflects organization model settings for the routines surface. Routines that run through an agent use the agent’s model instead.
- Review past runs — Open earlier runs to read outputs, timestamps, and errors when something failed.
- Stay in chat — Set up and tweak routines next to the threads where you already work.
Common patterns
- Daily summaries — inboxes, queues, or dashboards the agent can already reach with tools.
- Weekly reports — pull numbers or text from documents and connectors you trust.
- Monthly checks — same instructions you might type once in chat, repeated on a calendar.
How this fits with agents
You still shape the agent in Creating agents, then refine models and data in Configuring model and data sources and tools in Configure extended capabilities. A routine reuses that agent and adds when it runs and how tightly each run is scoped.
Organization-level routines (admins)
Organization admins and custom roles with organization-level Routines access can review scheduled work across the whole organization, not only their own routines.
Open the routines management page
- From the chat sidebar, open Routines.
- Use the Routines tab to see every routine you are allowed to view, and the Run history tab to filter executions by status, routine, and time range.
Members with only self-level Routines permission see routines they created. Members with organization-level Routines permission see routines created by anyone in the organization.
Agent and model precedence
Each routine runs against either an agent or a model:
| Target | What runs |
|---|---|
| Agent | The selected agent’s configuration—its model, tools, and instructions—drives each scheduled run. |
| Model | You pick the model (and tools) on the routine itself; the run does not inherit a full agent profile. |
The Agent column on the routines list shows which agent or model is configured so admins can spot conflicts—for example, two routines pointing at different agents for the same workflow.
When a routine targets an agent, the agent’s settings take precedence over ad hoc chat defaults. When it targets a model, the routine’s own model and tool selections apply.
Control who can view or edit routines
Grant Routines permissions under Organization > User Permissions when you create or edit a role:
- User (Self) — view or edit only that person’s routines.
- Organization — view or edit routines across the organization (typical for leads and admins auditing scheduled agents).
See Custom user permissions for how roles combine and how to assign them.
Related topics
- Getting started — sidebar navigation and how you open agents.
- Working with an agent — everyday chat next to scheduled runs.
- Agent tools and Add AI actions — how tools work before you attach them to a routine.
- App Builder — for publishing an app; use routines when you only need a scheduled agent, not a full app.
- Organization model settings — admin controls for which models appear in routines and other surfaces.