Extract structured Markdown (with tables, headings, and lists) and per-page rendered images from a PDF so Claude can read both modalities natively. Use this skill whenever the user asks to read, parse, summarize, analyze, OCR, transcribe, or "look inside" a PDF — local file or URL — and especially when the PDF contains figures, tables, equations, charts, scanned pages, multi-column layouts, or any visual content that plain text extraction would mangle. Strongly prefer this skill over writing one-off pdfminer / PyMuPDF / pdfplumber / pdf2image code.

2026-04-09

index-note

综合办公文员

统一知识管理 - 接受任意输入，自动分类整理为Obsidian笔记 / Unified knowledge management - accepts any input, auto-classifies and organizes into Obsidian notes

2026-04-09

index-chat

软件开发工程师

基于个性化人格与记忆库的智能聊天 / Persona-driven chat with memory-augmented retrieval from Obsidian vault

2026-04-07

index-init

软件开发工程师

初始化 Obsidian 知识库并配置 Agent 人格 / Initialize Obsidian vault and configure Agent persona

2026-04-07

index-update

软件开发工程师

整理已读笔记与聊天记录至分层知识库，归档至深度记忆 / Organize read notes and chat logs into hierarchical memory system, archive to deep memory

2026-04-07