MCP server for LLM fine-tuning with Unsloth. 33 tools, 180 tests, RunPod GPU integration. Fine-tune 2x faster with 80% less memory.
These files are ScientiaCapital/unsloth-mcp-server's own configuration. They tell Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.
CLAUDE.md A 378 tok .claude/skills/adaptive-workflows/SKILL.md A 51 tok .claude/skills/dataset-engineering/SKILL.md A 48 tok .claude/skills/model-deployment/SKILL.md B 52 tok .claude/skills/superbpe/SKILL.md A 58 tok .claude/skills/training-optimization/SKILL.md A 45 tok .claude/skills/unsloth-finetuning/SKILL.md A 66 tok .claude/skills/unsloth-tokenizer/SKILL.md A 53 tok .claude/commands/begin.md A 15 tok .claude/commands/build.md A 14 tok .claude/commands/done.md A 11 tok .claude/commands/end.md A 20 tok .claude/commands/fix.md A 14 tok .claude/commands/go.md A 16 tok .claude/commands/interrupt.md A 11 tok .claude/commands/pr.md A 14 tok .claude/commands/review.md A 6 tok .claude/commands/ship.md A 9 tok .claude/commands/simplify.md A 11 tok .claude/commands/status.md A 14 tok .claude/agents/observer-full.md A 39 tok .claude/agents/observer-lite.md A 51 tok