Skip to content

CLAUDE.md

Latest commit

 

History

History
48 lines (35 loc) · 2.06 KB

CLAUDE.md

File metadata and controls

48 lines (35 loc) · 2.06 KB
Raw

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Repository Overview

This is a documentation repository for the AI Learning Assistant (AI学习助手), a learning tool that integrates AI capabilities with Obsidian note-taking software. The repository contains user guides and documentation for different versions of the software.

Key Files and Structure

  • AI学习助手使用说明-软件版本V1.01.md - User guide for version 1.01, with images in images/ folder
  • AI学习助手1.1版本 使用文档.docx - Word document for version 1.1
  • AI学习助手1.1版本_使用文档.md - Converted markdown version of 1.1 documentation
  • images/ - Screenshots and diagrams for v1.01 documentation
  • media/ - Extracted images from Word document conversion (when using pandoc --extract-media)

Common Tasks

Converting Word Documents to Markdown

Use pandoc with media extraction to preserve images:

pandoc "filename.docx" --extract-media=. -o "output.md"

Image Path Management

  • V1.01 documentation uses images/ folder with descriptive Chinese filenames
  • Converted Word docs create media/ folder with generic image1.png, image2.png names
  • When merging documents, ensure image paths are consistent

Documentation Merging

When combining multiple markdown files:

  1. Align image folder structures (images/ vs media/)
  2. Ensure heading levels are consistent
  3. Maintain table of contents structure used in existing docs
  4. Preserve Obsidian-specific syntax like ![[filename.png]] and table-of-contents blocks

Content Structure

The documentation covers:

  • AI plugin usage (Copilot integration, custom personas)
  • Voice recognition (ASR) deployment and configuration
  • Text-to-speech (TTS) setup
  • Multi-platform deployment (Windows CPU/CUDA, macOS)
  • Obsidian plugin configuration and workflows

Language and Localization

All documentation is in Chinese. The software supports both Windows and macOS platforms with different deployment packages for CPU and CUDA environments.