git
ask
hub
Privacy
Terms
Sign in with GitHub
mrwwk/deskcraft
↗
Stars ·
119
Language ·
Python
License ·
Apache-2.0
Simplify and Visualize This Repo
Scan the safety of this repo
Self Host this repo
Check out more work of Developer: mrwwk
Ask anything about this repo to start.
I want to run my AI agent against the DeskCraft benchmark. Walk me through downloading the Ubuntu VM image from Hugging Face, setting up the Docker desktop provider, and running the evaluation scripts to get a score.
My agent completed a DeskCraft LibreOffice Calc task but the evaluation script says it failed. Here is the task description: [paste task]. What file state or export is the script checking, and how do I debug why it did not pass?
I want to add a new GIMP image-editing task to DeskCraft that requires exporting a specific layer. What format does the task definition use and how do I register the automatic success check?
Explain how DeskCraft's interactive collaboration tasks work. How are the mid-session user interrupts triggered, and what must my agent do differently compared to single-pass tasks to handle them correctly?
Full explanation on explaingit →
📎
Send
By chatting or signing in you agree to the
Terms
and chat-message logging (revocable in
History
).