A sandboxed CLI evaluation platform that measures not just what you solved — but how.
for now run cli with:
docker ps -a | grep clice | awk '{print $1}' | xargs -r docker rm -f && docker volume ls | grep clice-workspace | awk '{print $2}' | xargs -r docker volume rm -f && python clice.py run hello-cliceand run textual tui with:
python -m ui.main
Note
Now you would need to run both the tui and the cli in a linux or with wsl Current design allows that if there is an interruption during a session and you comeback it continues. however if you pass or fail (Basically get to the verdict screen, your progress is reset.)