AI-native interactive canvas for software project visualization and management.
The canvas IS the UI. Built for developers who think in graphs, not lists.
- Node.js 20+
- Python 3.12+
- pnpm (installed automatically via
npm install -g pnpm) - Docker (optional, for containerized development)
# 1. Clone and navigate
cd vislzr-unified
# 2. Install dependencies
pnpm install
# 3. Setup Python backend
cd packages/api
python3 -m venv .venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
pip install -r requirements.txt
# 4. Configure environment
cp .env.example .env
# Edit .env and add your GOOGLE_API_KEY (optional for AI features)
# 5. Start backend (in packages/api/)
uvicorn app.main:app --reload --port 8000
# 6. Start frontend (in new terminal, from root)
cd packages/web
pnpm dev
# 7. Open browser
# Frontend: http://localhost:5173
# API Docs: http://localhost:8000/docs# Start all services
docker-compose up
# Frontend: http://localhost:5173
# Backend: http://localhost:8000
# PostgreSQL: localhost:5432- PRD-MASTER.md - Product vision & complete feature spec
- PHASE-0-AUDIT.md - Codebase audit & migration assessment
- IMPLEMENTATION-PLAN.md - 12-week development roadmap
- AGENT-SDK-SETUP.md - Agent-based development workflow
- CLAUDE-CODE-GUIDE.md - Using Claude Code CLI
- ROADMAP-VISUAL.md - Visual development timeline
vislzr-unified/
โโโ docs/ # Documentation hub
โ โโโ PRD-MASTER.md # Product requirements (25k+ words)
โ โโโ PHASE-0-AUDIT.md # Migration audit & findings
โ โโโ IMPLEMENTATION-PLAN.md # 12-week roadmap
โ โโโ AGENT-SDK-SETUP.md # Agent coordination guide
โ โโโ sprints/ # Weekly sprint planning
โโโ packages/
โ โโโ api/ # FastAPI backend
โ โ โโโ app/ # Application code
โ โ โ โโโ main.py # FastAPI app entry
โ โ โ โโโ models.py # SQLAlchemy models
โ โ โ โโโ routes.py # API endpoints
โ โ โ โโโ ws.py # WebSocket handlers
โ โ โ โโโ ai_service.py # AI integration (Gemini)
โ โ โโโ tests/ # Backend tests
โ โ โโโ requirements.txt # Python dependencies
โ โโโ web/ # React frontend
โ โ โโโ src/
โ โ โ โโโ components/ # React components
โ โ โ โ โโโ GraphView.tsx # Main canvas (D3.js)
โ โ โ โ โโโ SidePanel.tsx # Node editor
โ โ โ โ โโโ Timeline.tsx # Milestone tracker
โ โ โ โ โโโ AIPrompt.tsx # AI generation UI
โ โ โ โ โโโ ...
โ โ โ โโโ services/ # API clients
โ โ โ โโโ App.tsx # Main app
โ โ โโโ package.json
โ โโโ shared/ # Shared TypeScript types
โ โโโ src/types/index.ts # Node, Edge, Graph types
โโโ agents/ # Agent context files
โ โโโ architect/ # Architecture planning
โ โโโ developer/ # Implementation work
โ โโโ security/ # Security reviews
โ โโโ qa/ # Testing & quality
โโโ docker-compose.yml # Multi-service orchestration
โโโ pnpm-workspace.yaml # PNPM monorepo config
โโโ turbo.json # Turborepo build config
- Monorepo structure created
- Backend migrated from Vislzr-main
- Frontend migrated with all components
- Shared types package created
- Docker compose setup
- Dependencies installed
- PRD gap analysis documented
Weeks 2-3: Implement canvas-centric interaction paradigm
Key Features:
- Sibling node rendering (D3.js)
- Context-aware action registry
- Animated transient UI elements
- 80% of actions moved from sidebar to canvas
See IMPLEMENTATION-PLAN.md Phase 1 for details.
- React 19 + TypeScript 5.8
- D3.js 7.9 (force-directed graphs)
- TailwindCSS 4 (styling)
- Vite 7 (build tool)
- Vitest (testing)
- FastAPI 0.115 (Python web framework)
- SQLAlchemy 2.0 (ORM)
- WebSocket (real-time updates)
- Google Gemini API (AI generation)
- pytest (testing)
- PNPM (package management)
- Turborepo (monorepo builds)
- Docker Compose (local dev)
- PostgreSQL (production DB)
- SQLite (dev/test DB)
- Interactive D3.js force-directed graph
- Node & edge CRUD operations
- Real-time WebSocket updates
- Timeline with milestones
- AI project generation (Google Gemini)
- JSON import/export
- Node styling (status colors, priority sizes)
- Sibling Nodes - Canvas-centric actions
- Context-aware UI - Intelligent action suggestions
- Enhanced node types - 14 node types per PRD
- Dependency focus mode & advanced visualizations
- AI scans (security, optimization, architecture)
- Contextual AI assistant on every node
- In-canvas code viewer/editor
- Git integration & file sync
- External integrations (GitHub, Docker, Cloud)
cd packages/api
source .venv/bin/activate
pytestcd packages/web
pnpm testpnpm test# Architecture planning
aider --architect --read docs/PRD-MASTER.md --message "Plan Phase 1"
# Implementation
aider --developer --file packages/web/src/components/SiblingNodes.tsx
# Testing
aider --qa --message "Create tests for sibling nodes"
# Security review
aider --security --message "Review today's changes"# Start backend
cd packages/api && uvicorn app.main:app --reload
# Start frontend (new terminal)
cd packages/web && pnpm dev
# Run tests
pnpm test
# Build all packages
pnpm build
# Lint
pnpm lint# Check Python version
python3 --version # Should be 3.12+
# Reinstall dependencies
cd packages/api
pip install --force-reinstall -r requirements.txt
# Check DATABASE_URL in .env
cat .env | grep DATABASE_URL# Clear node_modules and reinstall
rm -rf node_modules packages/*/node_modules
pnpm install
# Check Node version
node --version # Should be 20+# Reset SQLite database
rm packages/api/vislzr.db
# Reinitialize
cd packages/api && python
>>> from app.db import init_db
>>> init_db()This is a personal project currently in active development (Phase 0 โ Phase 1).
Development Phases:
- Phase 0 (Week 1): โ Foundation complete
- Phase 1 (Weeks 2-3): Sibling Nodes
- Phase 2 (Weeks 4-5): Advanced Visualizations
- Phase 3 (Weeks 6-8): AI Deep Integration
- Phase 4 (Weeks 9-10): Code Integration
- Phase 5 (Weeks 11-12): Polish & Launch
See IMPLEMENTATION-PLAN.md for detailed tasks.
MIT License
- Canvas-Centric UX - The visualization IS the interface
- Sibling Nodes - Transient, context-aware actions on canvas
- AI-Native - AI integrated throughout, not bolted on
- Developer-First - Built for how developers think
- Visual Intelligence - Color, size, animation convey meaning
Check docs/ directory for comprehensive guides
See docs/IMPLEMENTATION-PLAN.md for phase-specific guidance
Backend API docs: http://localhost:8000/docs (when running)
Built with โค๏ธ using React, TypeScript, D3.js, and FastAPI
Status: Phase 0 Complete โ | Next: Phase 1 (Sibling Nodes)
Last Updated: September 30, 2025