Summary
Cursor shipped Projects in beta to all users: a container for larger bodies of work (a feature, a migration, a full app) that maintains context over months. A Coordinator agent plans work, delegates to parallel implementing agents, and brings finished work back for review; it does not write code itself. A Project runs on its own cloud computer (closing the laptop does not stop it), launching local agents on demand for local testing. Files are synced across all of a project's cloud and local machines, accumulating research, artifacts, codebase knowledge and user preferences — once one agent learns how to test a service, every future agent reuses it. Coordinators can watch Slack channels, run on schedules, or follow PRs and act on signals.
Why it matters
Cursor has productized the manager/implementer split: the coordinator is a planning-and-delegation layer over the parallel agent fleet it already shipped, and the shared, growing context file turns one-off agent runs into a durable engineering asset. Combined with subscriptions, this is the most complete 'always-on engineering org' shipped by a dev-tool vendor — the open question is review load and cost control.
Technical details
| Date | 2026-09-10, beta rolling out to all users |
|---|---|
| Coordinator | plans, delegates to parallel agents, returns finished work for review; does not write code |
| Execution | Project runs on its own cloud computer; local agents launched on demand for local testing |
| Shared Context | files synced across all project machines; agents contribute research, artifacts, codebase knowledge, preferences; knowledge accumulates and is reused |
| Subscriptions | watch Slack channels / run on schedules / follow PRs; act on signals without prompting |
Tags
cursormulti-agentcoordinatorcloud-agentslong-running-agents