do-knowledge-studio

A local-first knowledge management studio — rich notes, knowledge graph, mind maps, full-text search, and AI agent integration in a single browser-based app.

Quick Links: Quick Start · Features · Architecture · AI Agents · Contributing

---

What Is This?

do-knowledge-studio is a local-first, offline-capable knowledge management application built with React, TypeScript, and SQLite WASM. It combines a rich-text editor, an interactive knowledge graph, mind mapping, blazing-fast full-text search, and a static site export — all running entirely in the browser with no backend required.

It also ships with a production-ready AI agent harness supporting Claude Code, Gemini CLI, OpenCode, Qwen Code, Windsurf, and Cursor — making it a great base for AI-assisted personal knowledge management workflows.

---

✨ Features

• 📝 Rich Text Editor — TipTap-powered WYSIWYG editor with Markdown support and placeholder hints
• 🗄️ Local SQLite Database — Persistent, offline-first storage via @sqlite.org/sqlite-wasm with FTS5 full-text search
• 🔍 Semantic & Full-Text Search — Orama-powered in-browser search index for fast, context-aware retrieval
• 🕸️ Knowledge Graph — Interactive node-link graph built with Graphology + Sigma.js; focus mode for deep neighborhood exploration
• 🧠 Mind Maps — Visual mind mapping with Mind Elixir for brainstorming and concept structuring
• 📤 Static Site Export — Turn your knowledge base into a shareable, self-contained static site
• 🤖 Multi-Agent AI Harness — Unified workflow across 6+ AI coding tools with skills, quality gates, and sub-agent patterns
• 🔬 Swarm Analysis — Parallel AI-powered web research using git worktrees
• ⚡ CLI — TypeScript CLI for scripting and automation tasks
• 🧪 Full Test Suite — Vitest unit tests + Playwright end-to-end tests

---

🚀 Quick Start

Prerequisites

• Node.js 20+ (install)
• A modern browser (Chrome, Firefox, Edge)
• (Optional) An AI coding CLI: Claude Code, Gemini CLI, OpenCode, Qwen Code

Installation

git clone https://github.com/d-oit/do-knowledge-studio.git
cd do-knowledge-studio
npm install

Development

npm run dev

Open http://localhost:5173 in your browser.

Build

npm run build
npm run preview

Environment

cp .env.example .env
# Edit .env as needed

---

🏗️ Architecture

src/
├── app/          # React app shell, routing, layout
├── db/           # SQLite WASM database layer + FTS5 search
├── features/     # Feature modules (editor, graph, mindmap, search, export)
├── lib/          # Shared utilities and Orama search index
└── main.tsx      # Entry point

cli/              # TypeScript CLI for automation
export/           # Static site export engine
scripts/          # Setup and quality gate scripts
tests/            # Playwright e2e tests

Tech Stack

Layer	Technology
UI Framework	React 18 + TypeScript 5
Build Tool	Vite 8
Database	SQLite WASM (FTS5)
Search	Orama 3
Rich Text	TipTap 2
Graph	Graphology + Sigma.js
Mind Map	Mind Elixir 5
Validation	Zod
Icons	Lucide React
Unit Tests	Vitest 4
E2E Tests	Playwright

---

🤖 AI Agent Integration

This project ships with a unified AI agent harness that works across multiple tools:

AGENTS.md           # Single source of truth for all agents
├── CLAUDE.md       # Claude Code overrides
├── GEMINI.md       # Gemini CLI overrides
├── QWEN.md         # Qwen Code overrides
└── opencode.json   # OpenCode configuration

Skills System

Reusable knowledge modules live in .agents/skills/ and are symlinked into each tool's config directory:

.agents/skills/         # Canonical skill source
.claude/skills/         # Symlinks → ../../.agents/skills/
.gemini/skills/         # Symlinks → ../../.agents/skills/
.qwen/skills/           # Symlinks → ../../.agents/skills/

Setup Agent Skills

# Create skill symlinks (run once)
./scripts/setup-skills.sh

# Validate setup
./scripts/validate-skills.sh

Sub-Agent Patterns

graph LR
    A[Main Agent] --> B[Sub-Agent 1]
    A --> C[Sub-Agent 2]
    B --> D[Task Complete]
    C --> E[Task Complete]
    D --> F[Synthesize]
    E --> F

Loading

---

🧪 Testing

# Unit tests
npm run test

# Watch mode
npm run test:watch

# End-to-end tests
npm run test:e2e

# Type checking
npm run typecheck

# Lint
npm run lint

---

⚙️ CLI

npm run cli -- <command>

See cli/ directory for available commands and scripting options.

---

📚 Documentation

• 📖 AGENTS.md — AI agent instructions (single source of truth)
• ⚡ QUICKSTART.md — 5-minute setup guide
• 🏗️ Harness Overview — Architecture and patterns
• 🪝 Skills Guide — Creating reusable skills
• 🤖 Sub-Agents — Context isolation patterns
• 🔧 Hooks — Pre/post tool hooks
• 📊 Context — Back-pressure mechanisms
• ⚙️ Configuration — Repository constants and utilities
• 🔄 Migration — Adopting in existing projects
• 🔒 Security — Security policy and reporting
• 📝 Changelog — Release history

---

🤝 Contributing

Contributions are welcome! See CONTRIBUTING.md for:

• Development environment setup
• Code style and testing requirements
• Pull request process
• Good first issues

Community

• 🐛 Issue Tracker — Report bugs
• 💬 Discussions — Ask questions

---

License

Licensed under the MIT License.

---

Local-first. AI-assisted. Built to think.