Nilay cf0b66c83a Studio: keep skills created in Docker on the unsloth-studio volume (#13912) ... In the Studio Docker images, a skill made from the Skills dialog or by the model's create_skill tool was saved under /root/.agents/skills, which is inside the container and not on the unsloth-studio volume. After `docker rm` and a new `docker run` with the same volume, the documented way to move to a newer image, every created skill was gone with no warning, while accounts, chats and project folders came back. The Studio images now save the owner's skills in /opt/unsloth-studio/skills on the volume, next to project folders and the same layout other accounts already use for their own skills. They are listed, edited and deleted there, and the dialog shows that path. Installs outside Docker keep using ~/.agents/skills, and ~/.claude/skills is still read as before. The Docker Hub pages now list skills among the data the volume keeps. Skills made by an image from before this change stay in that old container's /root/.agents/skills, so copy them into /opt/unsloth-studio/skills before removing it.		2026-10-18 05:50:30 -07:00
..
jupyter	Bump Desktop crates, Docker JupyterLab, setuptools and CI pins for advisories (#12704)	2026-10-04 17:19:09 -07:00
.dockerignore	Docker: allow unsloth_root_shim.py into the build context (#12929)	2026-10-07 01:52:13 -07:00
build.sh	Keep the Docker notebooks and the models they save after docker rm (#13660)	2026-10-14 10:13:52 -07:00
Dockerfile	Keep the notebook and saved model when unsloth-run runs a URL (#12907)	2026-10-07 20:13:49 +13:00
Dockerfile.rocm	Lift the transformers ceiling to 5.17.0 and align the mirrored CI caps (#11016)	2026-10-02 07:27:45 -07:00
Dockerfile.studio	Studio: keep skills created in Docker on the unsloth-studio volume (#13912)	2026-10-18 05:50:30 -07:00
Dockerfile.studio-rocm	Studio: keep skills created in Docker on the unsloth-studio volume (#13912)	2026-10-18 05:50:30 -07:00
DOCKERHUB-ROCM.md	Studio: keep skills created in Docker on the unsloth-studio volume (#13912)	2026-10-18 05:50:30 -07:00
DOCKERHUB.md	Studio: keep skills created in Docker on the unsloth-studio volume (#13912)	2026-10-18 05:50:30 -07:00
entrypoint-rocm.sh	Docker ROCm Studio image: ship JupyterLab with the notebooks, as the CUDA image does (#11286)	2026-09-21 07:00:07 -07:00
entrypoint.sh	Use Unsloth branding in remaining user-facing text (#11254)	2026-09-18 01:58:29 -07:00
fetch_llama_prebuilt.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
install_nvidia_toolkit.sh	docker: stop the toolkit installer reporting a healthy driver as missing (#11106)	2026-09-16 05:00:38 -07:00
NOTICE	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
run.sh	Keep the Docker notebooks and the models they save after docker rm (#13660)	2026-10-14 10:13:52 -07:00
smoke_test.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
smoke_test_rocm.py	Keep the Docker notebooks and the models they save after docker rm (#13660)	2026-10-14 10:13:52 -07:00
studio_home.sh	Studio: keep skills created in Docker on the unsloth-studio volume (#13912)	2026-10-18 05:50:30 -07:00
studio_launch.sh	Docker ROCm Studio image: ship JupyterLab with the notebooks, as the CUDA image does (#11286)	2026-09-21 07:00:07 -07:00
studio_launch_rocm.sh	Docker ROCm image: ship Unsloth Studio, as the CUDA image already does (#11218)	2026-09-20 06:05:21 -07:00
studio_password.sh	Use Unsloth branding in remaining user-facing text (#11254)	2026-09-18 01:58:29 -07:00
studio_run.sh	Docker: let UNSLOTH_STUDIO_PORT change Studio's port inside the container (#11045)	2026-09-16 06:28:15 -07:00
supervisord.conf	Docker ROCm Studio image: ship JupyterLab with the notebooks, as the CUDA image does (#11286)	2026-09-21 07:00:07 -07:00
test_locally-rocm.sh	feat(docker): add an AMD ROCm image, for RDNA2 through RDNA4 and CDNA (#10820)	2026-09-14 06:06:59 -07:00
unsloth_colab_compat.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
unsloth_ipython_startup.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
unsloth_jupyter_tunnel.sh	Docker: keep the startup health checks off the HTTP proxy (#11157)	2026-09-17 06:28:58 -07:00
unsloth_llama_update.sh	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
unsloth_nb_compat.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
unsloth_nb_content_sig.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
unsloth_nb_pip_magic.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
unsloth_nb_strip_colab.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
unsloth_nb_view.py	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00
unsloth_pip_shim.py	Keep the notebook and saved model when unsloth-run runs a URL (#12907)	2026-10-07 20:13:49 +13:00
unsloth_root_shim.py	Keep the notebook and saved model when unsloth-run runs a URL (#12907)	2026-10-07 20:13:49 +13:00
unsloth_run.py	Keep the notebook and saved model when unsloth-run runs a URL (#12907)	2026-10-07 20:13:49 +13:00
unsloth_studio_update.sh	Probe Studio before reading the deadline in unsloth-studio-update (#11590)	2026-09-22 15:12:56 -07:00
unsloth_sync_notebooks.sh	Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)	2026-09-04 16:26:55 -07:00