Karpathy Compile OpenClaw Skill
Compile raw wiki entries from Phase 1 into structured, distilled knowledge points using LLM, grouping by topic and saving refined outputs.
Installation
clawhub install karpathy-compile
Requires npm i -g clawhub
58
Downloads
0
Stars
0
current installs
0 all-time
1
Versions
EU-Hosted Inference API Power your OpenClaw skills with
the best open-source models.
Drop-in OpenAI-compatible API. No data leaves Europe.
Explore Inference APIGLM
GLM 5
$1.00 / $3.20
per M tokens
Kimi
Kimi K2.5
$0.60 / $2.80
per M tokens
MiniMax
MiniMax M2.5
$0.30 / $1.20
per M tokens
Qwen
Qwen3.5 122B
$0.40 / $3.00
per M tokens
Karpathy Compile Skill - Phase 2
描述
实现 Karpathy LLM Knowledge Base 的第二阶段：Wiki → Knowledge Points 编译。
将 Phase 1 生成的 wiki 条目通过 LLM distillation 编译为结构化的知识精华（knowledge points）。
工作流程
Phase 1: 用户查询 → Wiki 条目 (raw, 多条)
Phase 2: Wiki 条目 → Knowledge Points (精炼, 结构性)
Phase 3: Lint → 去重/合并/更新Knowledge Point 格式
## Knowledge Point: [主题]
**核心概念**: [一句话概括]
**来源**: [wiki条目来源]
**详细说明**: [LLM生成的详细解释]
**标签**: [tag1, tag2]
**创建时间**: YYYY-MM-DD
**可信度**: high/medium/lowCompile Pipeline
- 读取 wiki 文件
- 按主题/标签分组
- 对每组使用 LLM distillation 生成 knowledge point
- 保存到 knowledge-points/ 目录
文件结构
karpathy-compile/
├── SKILL.md
├── scripts/
│ ├── __init__.py # CompilePipeline
│ ├── distiller.py # LLM distillation
│ ├── parser.py # wiki 文件解析
│ └── test_compile.py # 测试
└── knowledge-points/ # 输出目录依赖
- Phase 1 的 wiki 文件
- M-Flow (用于存储编译后的 knowledge points)
- Ollama LLM (qwen2.5:14b)
Statistics
Author
sune
@sora-mury
Latest Changes
v1.0.0 · Apr 5, 2026
Initial release: Wiki to Knowledge Points compilation pipeline
Quick Install
clawhub install karpathy-compile Related Skills
Other popular skills you might find useful.
Made in Europe Chat with 100+ AI Models in one App.
Use Claude, ChatGPT, Gemini alongside with EU-Hosted Models like Deepseek, GLM-5, Kimi K2.5 and many more.
