ThinkWiki

Initialize and maintain a local Markdown knowledge base with automated inbox management.

30|5|Updated Jun 20, 2026
One-click install
npx skills add https://github.com/wzdavid/ThinkWiki --skill thinkwiki
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ThinkWiki
Source: https://github.com/wzdavid/ThinkWiki/tree/main
Command: npx skills add https://github.com/wzdavid/ThinkWiki --skill thinkwiki

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires markitdown, beautifulsoup4, markdownify, mammoth, cryptography, pdfminer-six, pdfplumber, pypdfium2, openpyxl, pandas, python-pptx, lxml, xlrd, and includes scripts (resource) components.

What problem does it solve?

ThinkWiki solves the problem of fragmented information by transforming scattered documents, web pages, and conversations into a structured, searchable, and browsable local Markdown knowledge base.

Core Features & Use Cases

  • Inbox-First Capture: Collect web clips, files, and notes into a dedicated inbox for review before formal ingestion.
  • Knowledge Graph Generation: Automatically build an interactive content knowledge graph that maps concepts, entities, and semantic relations.
  • Evidence-Based Q&A: Query your wiki to receive answers backed by specific evidence from your own documents.
  • Use Case: A researcher can clip multiple web articles into the inbox, ingest them into the wiki, and then ask the agent to generate a knowledge graph to visualize the relationships between the research topics.

Quick Start

Ask your agent to create a new local knowledge base named Research Wiki in your current workspace.

Frequently Asked Questions about ThinkWiki

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a local Markdown knowledge base from scattered notes and web clippings?

To build a local Markdown knowledge base, you can collect web clips, files, and notes into a dedicated inbox for review before ingesting them into a structured, searchable wiki workspace.

Can I generate a knowledge graph to visualize relationships between entities in my research documents?

Yes, you can generate an interactive knowledge graph to visualize relationships by automatically mapping concepts, entities, and semantic relations from your ingested Markdown documents.

How do I query my personal wiki and get answers backed by evidence from my own documents?

You can query your personal wiki using evidence-based question answering, which retrieves answers backed by specific evidence extracted directly from your local Markdown documents.

What do I need to set up a local Markdown wiki workspace with knowledge graph visualization?

You need Python 3 and local file system access to initialize the wiki workspace, manage the automated inbox, and generate the HTML-based navigation artifacts for knowledge graph visualization.

Does this knowledge base tool support ingesting files like PDFs, Excel spreadsheets, and PowerPoint presentations?

Yes, the knowledge base supports document ingestion from various formats including PDFs, Excel spreadsheets, and PowerPoint presentations using dependencies like pdfplumber, openpyxl, and python-pptx.