Cursor engineer @mattyp shows a Mac Mini becoming a Cloud Agent worker via agent worker start; --computer-use enables remote click/type/screenshot through Cursor Computer Use. Official Self-Hosted Machines keep inference in Cursor cloud and tool execution on your hardware over outbound HTTPS (My Machines + Team Pools).

Key Takeaways

  • ✓Product path: Self-Hosted Machines blog + docs — tools on your hardware, loop in Cursor cloud
  • ✓Personal quickstart (My Machines): install CLI → agent login → agent worker start (outbound HTTPS only)
  • ✓Desktop control: agent worker --computer-use start; grant Accessibility + Screen Recording to Cursor Computer Use on macOS
  • ✓Scale: 200 workers/user, 1000/team; Team Pools need Enterprise + service-account API key
  • ✓Verify with agent worker debug + screenshot task; route via worker= from Slack/GitHub/Linear
Cursor Self-Hosted: one `agent worker start` turns a Mac Mini into a remote worker; `--computer-use` drives the desktop
🖼️Official Media
Click to view high-res
🧭

Not enough VRAM? Compare cloud API and self-hosting costs

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

🔬

In-Depth Technical Analysis

Core background

Managed Cloud Agents run in Cursor-hosted VMs. Teams that need private SCM, custom GPU/Mac (iOS), or existing build pipelines can move tool execution onto hardware they own via Self-Hosted Machines — while inference/planning stay in Cursor's cloud. Cursor engineer @mattyp highlighted the personal path: one CLI line turns a Mac Mini into a remote worker, optionally with Computer Use.

Architecture

agent worker start opens a long-lived outbound HTTPS session to api2.cursor.sh (no inbound ports). My Machines for personal boxes; Team Pools (Enterprise + service-account key) for fleets. --computer-use installs the macOS helper Cursor Computer Use on first start (or X11/TigerVNC on Linux). Caps: 200 workers/user, 1000/team. Docs: self-hosted overview, My Machines, Computer use. Blog: Run cloud agents on machines you manage (2026-09-02).

Benchmarks

No official Terminal-Bench / SWE-Bench (or latency) numbers for this product path. Validate with agent worker debug plus a real screenshot/click task after granting Accessibility + Screen Recording to Cursor Computer Use.

Developer guide

Install CLI → agent login → agent worker --computer-use --name "mac-mini" start. Keep a signed-in GUI session awake. Pick the machine at cursor.com/agents; route via worker= from Slack/GitHub/Linear.

Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.

Up NextCursor launches Computer Use: agents can control your computer and drive desktop apps like the iOS Simulator to test what they build
Read Now→