See how a minor change to your commit message style can make a difference.

git commit -m"<type>(<optional scope>): <description>" \
  -m"<optional body>" \
  -m"<optional footer>"

A pattern for building personal knowledge bases using LLMs.

This is an idea file, it is designed to be copy pasted to your own LLM Agent (e.g. OpenAI Codex, Claude Code, OpenCode / Pi, or etc.). Its goal is to communicate the high level idea, but your agent will build out the specifics in collaboration with you.

Most people's experience with LLMs and documents looks like RAG: you upload a collection of files, the LLM retrieves relevant chunks at query time, and generates an answer. This works, but the LLM is rediscovering knowledge from scratch on every question. There's no accumulation. Ask a subtle question that requires synthesizing five documents, and the LLM has to find and piece together the relevant fragments every time. Nothing is built up. NotebookLM, ChatGPT file uploads, and most RAG systems work this way.

使用JAV金鸡儿奖官网附带的工具JAV SQL 查询器，可查询各种类别的JavDB TOP250影片：

• JavDB TOP250
• JavDB 有码 TOP250
• JavDB 无码 TOP250
• JavDB 欧美 TOP250
• JavDB FC2 TOP250

及分年数据（存在部分重复影片，原始数据的问题）：

Getting Started with Claude Code — A practical guide for knowledge workers, not engineers

A practical guide for getting started with Claude Code — written for knowledge workers, not engineers. By Melody Koh (Ground Truth), updated April 2026.

Note: This guide is for Claude Code (the CLI/desktop tool). If you use Cowork (the web/desktop app), the core concepts (context, memory, skills) translate — but the setup and commands are different. See the Cowork vs Claude Code comparison guide to understand the tradeoffs, or Anthropic's Cowork getting started guide for setup help.