Knowledge Graph Kit

Build, visualize, and explore knowledge graphs with an interactive web interface. Perfect for research synthesis, systems mapping, or stakeholder analysis.

Live demo — an interactive ecosystem map of the author's public projects, built with this kit.

Quick Start • Templates • Documentation

---

Features

• 🎯 4 Pre-built Templates - Research, Systems Mapping, Ecosystem Mapping, Generic
• 🎨 Interactive Visualization - Click, explore, and navigate your knowledge graph
• 🤖 AI Integration - Optional Gemini chat for insights and exploration
• 📊 Provenance Tracking - Track where each concept came from
• 🔧 Fully Configurable - Customize entity types, relationships, and visuals via YAML
• 💻 Simple Setup - Interactive wizard walks you through everything

---

Quick Start

Interactive Setup (Recommended)

cd knowledge-graph-kit
python setup_wizard.py

The wizard will:

1. Help you choose a template
2. Configure AI integration (optional)
3. Set up your project
4. Launch the viewer

Takes 2-5 minutes.

Manual Setup

# 1. Choose and copy a template
cp -r templates/research ./my-project

# 2. Install dependencies
pip install -r requirements.txt

# 3. Initialize
cd my-project
python init.py

# 4. Start the server
python server.py  # Basic
# or
python start_server.py  # With AI chat

Open http://localhost:8000/viewer.html in your browser.

---

Templates

Choose the template that fits your use case:

🎓 Research

For: Academic papers, literature reviews, systematic analysis

Entities: Concepts, Researchers, Papers
Use Cases: Literature synthesis, concept mapping, citation networks

🏗️ Systems Mapping

For: Systems thinking, complex systems analysis, organizational systems

Entities: Components, Teams, Specifications
Use Cases: Systems thinking, dependency mapping, impact analysis, systems documentation

🌐 Ecosystem Mapping

For: Stakeholder analysis, value flows, organizational networks

Entities: Elements, Stakeholders, Insights
Use Cases: Ecosystem analysis, power mapping, value chain visualization

⚙️ Generic

For: Custom domains not covered above

Entities: Fully customizable
Use Cases: Anything else - you define the schema

See full template comparison →

---

Example Usage

Quick Start Code

from core.graph_manager import GraphManager

# Initialize
gm = GraphManager('config.yaml')

# Add entities
gm.add_entity('primary', {
    'id': 'design-thinking',
    'label': 'Design Thinking',
    'type': 'methodology',
    'description': 'User-centered, iterative innovation process',
    'source_papers': ['paper-001']
})

# Add relationships
gm.add_relationship('design-thinking', 'human-centered-design', 'relates-to')

# Save
gm.save()

📊 Complete Working Example

See a real systems map of the Knowledge Graph Kit codebase itself:

cd examples/systems-map-example
python build_map.py
python server.py
# Opens at http://localhost:8000/viewer.html

Includes: 20 components, 39 relationships, full documentation

View example documentation →

---

Viewer Features

Feature	Description
Interactive Graph	Drag, zoom, and explore nodes
Smart Search	Find entities by name or description
Click-through Navigation	Navigate between related entities
Show Connections	Focus on subgraphs
Breadcrumb Trail	Track your exploration path
Document Panel	View source documents inline
AI Chat	Ask questions about your graph (optional)

---

Project Structure

your-project/
├── config.yaml              # Configuration
├── _data/
│   └── entities.json        # Your knowledge graph
├── viewer.html              # Interactive visualization
├── server.py                # HTTP server
└── README.md                # Template docs

---

Configuration

All templates use YAML configuration:

domain: research

entity_types:
  primary:
    name: concepts
    types: [methodology, framework, technology]
  
  contributors:
    name: researchers
  
  sources:
    name: papers

relationships:
  types: [extends, builds-on, challenges]

visualization:
  colors:
    methodology: "#3498db"
    framework: "#9b59b6"

See the Generic template guide for all config options →

---

AI Integration (Optional)

Enable Gemini AI chat for:

• Asking questions about your graph
• Finding patterns and connections
• Getting synthesis insights

Setup:

1. Get API key from Google AI Studio
2. Run setup wizard and choose "Enable Gemini"
3. Or manually: copy gemini_config.template.json to gemini_config.json and add your key

---

Documentation

• Template Comparison - Detailed comparison of all templates
• Examples - Code examples for common tasks
• Template READMEs - In each template directory

---

Requirements

• Python 3.7+
• PyYAML
• google-generativeai (for AI features)

pip install -r requirements.txt

---

Use Cases

Academic Research

• Build cumulative knowledge from papers
• Track concept evolution across sources
• Identify research gaps
• Generate literature reviews

Systems Mapping & Thinking

• Map system dependencies and relationships
• Understand complex systems and feedback loops
• Analyze system dynamics and interactions
• Track ownership and specifications

Ecosystem Analysis

• Map stakeholder relationships
• Analyze power structures
• Trace value flows
• Identify collaboration opportunities

---

Contributing

Contributions welcome! Please feel free to submit a Pull Request.

---

License

MIT License - see LICENSE for details.

---

Acknowledgements

Built for researchers, architects, and analysts who need to understand complex, interconnected systems.