Apps · PROJECT NOTES
DeskCraft
An Ubuntu desktop benchmark for professional creative and engineering workflows, with 386 standard tasks and 152 tasks that add phased simulated-user interaction.
What can you do with it?
Evaluate long workflows across office, browser, design, video, audio and 3D applications using task-specific state and artifact checks. Interactive tasks test clarification and adaptation to later instructions through a user simulator.
What should you know before using it?
The published task guest is Ubuntu; inherited VM-provider code does not establish equivalent Windows or macOS task suites. Interactive replies use a model-backed user simulator, and optional completion triggers also use a model, so pin these configurations. Public VM and task-cache assets are separate downloads. The README recommends Python 3.10+, while package metadata requires 3.12+; some interactive examples retain older paths. Code and the cache card declare Apache-2.0, but the VM repository does not declare a license. This source review did not run the tasks.
What is verified here?
Runtime: Ubuntu guest with a compatible OSWorld-derived VM provider, public VM/cache assets and configured agent/user-simulator endpoints. Use the current standard_task.json and interactive_task.json indices with the matching ubuntu_examples base directory.
License note: Repository code is Apache-2.0; the task-cache card declares Apache-2.0. No license was stated for the separate Ubuntu VM repository; bundled applications and source assets retain their own terms.
This entry was reviewed against original project material. We have not independently run the software or reproduced its results. AI-development claims are author-reported; check upstream for current versions, setup requirements, and licenses.
- Resource type
- Environment + benchmark
- Provenance
- Development provenance not established
- Reviewed
- Platforms
- Linux