SwanLab gives you experiment tracking without the SaaS lock-in. It logs metrics, configs, and media (images, audio, point clouds) and works with PyTorch, Transformers, Lightning, and Fastai out of the box. You can run it locally, self-host, or use their cloud. The skill includes media dependencies pre-configured, so you can start logging visualizations immediately. If you're tired of sending training data to third-party platforms or need to track experiments in air-gapped environments, this is a solid pick. It's open source and the interface handles run comparisons across hyperparameters and seeds cleanly.
npx -y skills add orchestra-research/ai-research-skills --skill experiment-tracking-swanlab --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
sickn33/antigravity-awesome-skills
moizibnyousaf/ai-agent-skills
github/awesome-copilot