Playbooks overview
A playbook gives an AI agent a job to do. You write down the goal, list the outcomes (tasks) you want, and choose which actions the agent may run. When someone starts a run inside a space, the agent works through the tasks toward that goal. It reads files, searches the knowledge graph, and dispatches actions. It pauses for your approval at the gates you set.
Playbooks are org-level and reusable. A playbook is not tied to one space. You make it available to a space type. Every space created from that type can run it.
Playbooks (the definitions) are admin-only. The library lives at Playbooks in the sidebar. If you are not an admin, you see “You need an admin role to manage playbooks.” Running a playbook happens inside a space. Anyone who can contribute there can run one. See Run a playbook.
Two run types — keep them straight
A space has both a Playbooks tab and an Actions tab. They are not the same thing.
This page and the rest of this section cover playbook runs (the Playbooks tab). To run one action over some files, see Run an action.
The playbooks library
Open Playbooks to see every playbook in your organization. The list has these columns:
Use New playbook to start one. See Build a playbook for the editor walkthrough.
The lifecycle
Every playbook carries a status. It moves through three states.
Runs are always started by a person
There is no trigger or scheduling system. A playbook never runs on its own, on a timer, or in response to an event. A person starts every run from a space’s Playbooks tab.
Where to go next
Set the goal, add tasks, choose actions and approvals, then activate.
Start a run in a space and follow the live timeline.
Approve or reject the agent’s proposals, steer it mid-run, and control the run.
The installable work that a playbook’s agent, or you, can run over files.