serithemage/korea-ai-foundation-model-verification

A project verifying whether the publicly released models from the five Korean government AI foundation model institutions (Upstage, NAVER, SKT, NC, and LG) were actually trained from scratch

9 files for Claude Code: save, PostToolUse, korea-ai-foundation-model-verification, karpathy-guidelines and 5 more — 3,321 tokens loaded in every session.

Open on GitHub

Instructions

korea-ai-foundation-model-verification CLAUDE.md CLAUDE.md A 3,321 tok

Settings

korea-ai-foundation-model-verification .claude/settings.json A —

Hooks

PostToolUse .claude/settings.json A —

Skills

karpathy-guidelines .claude/skills/karpathy-guidelines/SKILL.md A —
llm-dna-extraction-playbook .claude/skills/llm-dna-extraction-playbook/SKILL.md A 154 tok
llm-dna-knowledge .claude/skills/llm-dna-knowledge/SKILL.md A 101 tok
update-changelog .claude/skills/update-changelog/SKILL.md A 49 tok
update-tutorial .claude/skills/update-tutorial/SKILL.md A 64 tok

Commands

save .claude/commands/save.md A —

These files are serithemage/korea-ai-foundation-model-verification'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.