An automated system that visualizes the relationships between CVE Numbering Authorities (CNAs) and Common Weakness Enumerations (CWEs) using interactive graph visualization.
This project automatically:
- Clones the CVE List V5 repository
- Parses all CVE JSON files to extract CNA-CWE associations
- Builds a NetworkX graph mapping these relationships
- Deploys multiple interactive visualizations to GitHub Pages
- Updates automatically every 3 hours via GitHub Actions
- Automated Data Processing: Python script parses 315,567 CVE records
- 10 Interactive Visualizations:
- Landing page with navigation
- Force-directed network graph
- Independent cluster view
- CWE co-occurrence network
- Top 50 CNA-CWE bipartite layout
- CWE hierarchy tree
- CWE star graphs
- Circular CWE layout
- CNA collaboration network
- MITRE ego network
- Interactive Controls: Pan, zoom, drag, search, and filter
- Auto-updating: GitHub Actions workflow runs every 3 hours
- Dark Mode UI: Modern, responsive design with Tailwind CSS
- Optimized Performance: 365-day data filter for faster loading
/
βββ .github/
β βββ workflows/
β βββ update_map.yml # GitHub Actions workflow
βββ scripts/ # Build and setup scripts
β βββ config.py # Configuration management
β βββ build_graph.py # Main graph generation script
β βββ build_alternative_graphs.py # Alternative visualizations
β βββ build_compact_graphs.py # Compact NetworkX layouts
β βββ build_extended_graphs.py # Extended visualizations (CVSS, vendors, etc.)
β βββ setup_cve_data.sh # CVE data repository setup
β βββ README.md # Scripts documentation
βββ web/ # Web visualizations
β βββ index.html # Landing page with navigation
β βββ force-directed.html # Force-directed graph
β βββ clusters.html # Independent clusters view
β βββ cwe-cooccurrence.html # CWE co-occurrence network
β βββ bipartite-top50.html # Top 50 CNA-CWE bipartite
β βββ cwe-hierarchy.html # CWE hierarchy tree
β βββ cwe-stars.html # CWE star graphs
β βββ cwe-circular.html # Circular CWE layout
β βββ cna-collaboration.html # CNA collaboration network
β βββ mitre-ego.html # MITRE ego network
β βββ data/ # Generated visualization data
β βββ cna_to_cwe_map.json # Main graph data
β βββ cwe_cooccurrence_map.json # CWE co-occurrence
β βββ top_cna_cwe_bipartite.json # Bipartite layout
β βββ cwe_hierarchy_tree.json # Hierarchy tree
β βββ cwe_star_graphs.json # Star graphs
β βββ cwe_circular_layout.json # Circular layout
β βββ cna_collaboration.json # Collaboration network
β βββ mitre_ego_network.json # MITRE ego network
βββ pyproject.toml # Python package configuration
βββ requirements.txt # Python dependencies
βββ .gitignore
βββ LICENSE
βββ README.md
- Clones the CVEListV5 repository
- Recursively walks through CVE JSON files
- Filters CVEs to those published in the last year (365 days) for performance
- Extracts CNA (
assignerOrgId) and CWE identifiers - Counts associations between CNAs and CWEs
- Builds a NetworkX graph with weighted edges
- Landing Page: Navigation hub with stats and links to visualizations
- Force-Directed Graph: Loads graph data using D3.js and renders interconnected layout
- Independent Clusters: Each CNA displayed as isolated mini-network
- Node colors: Blue (CNAs) vs Red (CWEs)
- Node sizes based on connection count
- Edge opacity/width based on association weight
- Interactive tooltips and selection highlighting
- Triggers every 3 hours (or manually)
- Sets up Python environment
- Installs NetworkX
- Clones fresh CVE data
- Runs build script
- Deploys to GitHub Pages
-
Clone the repository:
git clone https://github.com/RogoLabs/CVEMaps.git cd CVEMaps -
Install dependencies:
pip install networkx
-
Clone CVE data:
git clone --depth 1 https://github.com/CVEProject/cvelistV5.git cve-data
-
Build the graph:
python build_graph.py python build_alternative_graphs.py python build_compact_graphs.py
-
View locally: Open
web/index.htmlin your browser or use a local server:cd web python -m http.server 8000 # Visit http://localhost:8000
- Overview of all 10 available visualizations
- Live stats (CNAs, CWEs, associations, total CVEs)
- Quick navigation cards to different views
- Color-coded by visualization type
- All CNAs and CWEs in one interconnected graph
- 349 CNAs, 577 CWEs, 6,103 associations
- Drag nodes to explore relationships
- Click to highlight connections
- Real-time search filtering
- File size: 597KB
- 409 isolated mini-networks (one per CNA)
- CNA names displayed inside blue circles
- Sort by size or filter by name/CWE
- Adjustable density slider
- CNA size scales with CWE count
- Drag CWEs within cluster boundaries
- Optimized with 365-day filter
- Shows which CWEs commonly appear together in CVEs
- 577 CWE nodes, 903 co-occurrence edges
- Click to see common partners
- Node size reflects co-occurrence frequency
- File size: 121KB
- Two-column layout: 50 most active CNAs (left), 603 CWEs (right)
- 4,175 associations clearly visible
- Bipartite graph layout for pattern recognition
- Vertical divider separates the two groups
- File size: 444KB
- 3-level tree structure: root β categories β CWEs
- 51 nodes organized hierarchically
- Color-coded by type (blue=root, purple=category, red=CWE)
- Node size reflects CVE count
- File size: 14KB
- Top 12 CWEs displayed as star centers
- 252 total nodes with CNAs radiating outward
- Shows which organizations focus on which weaknesses
- Multiple stars in grid layout
- File size: 71KB
- Top 40 CWEs arranged in a symmetric circle
- 227 interconnection edges
- Click to highlight neighbors
- Perfect for spotting central weaknesses
- File size: 26KB
- 396 CNAs connected by shared CWE focus areas
- 4,190 collaboration edges
- Reveals research interests and patterns
- Spring layout with pre-computed positions
- File size: 521KB
- MITRE Corporation at center with direct connections
- 340 nodes at radius 1 from MITRE
- Radial layout from center node
- Red center node, blue connected CNAs
- File size: 75KB
All visualizations use:
- D3.js v7 for rendering and interactions
- NetworkX for graph generation and layout algorithms
- Pre-computed positions for instant loading (no client-side layout computation)
- Tailwind CSS for consistent styling
- Interactive features: pan, zoom, drag, click selection, search filtering
- Dark theme with gradient backgrounds
-
Enable GitHub Pages:
- Go to repository Settings β Pages
- Source: GitHub Actions
-
Run the workflow:
- Go to Actions tab
- Select "Update CVE Map Data"
- Click "Run workflow"
-
View your site:
- Visit
https://<username>.github.io/CVEMaps/
- Visit
The main visualization displays:
- 349 CNAs: CVE Numbering Authorities
- 577 CWEs: Common Weakness Enumerations
- 6,103 Associations: Unique CNA-CWE mappings
- 50,282 CVEs: From last 365 days (filtered from 315,567 total)
- Node Connections: Number of relationships per entity
- Edge Weights: Frequency of associations
- 365-day filter: Only processes CVEs from the last year for faster loading
- Pre-computed layouts: NetworkX calculates positions server-side
- Optimized file sizes: All visualizations under 600KB
- Efficient rendering: D3.js uses pre-computed coordinates
- Pan: Click and drag background
- Zoom: Mouse wheel or pinch
- Drag Nodes: Click and drag any node
- Select Node: Click to highlight connections
- Reset View: Button to reset pan/zoom
- Clear Selection: Remove highlighting
- Blue Nodes: CVE Numbering Authorities (CNAs)
- Red Nodes: Common Weakness Enumerations (CWEs)
- Node Size: Proportional to connection count
- Edge Thickness: Based on association weight
- Edge Opacity: Indicates relationship strength
Edit .github/workflows/update_map.yml:
schedule:
- cron: '0 */3 * * *' # Every 3 hoursEdit build_graph.py force simulation settings:
- Node repulsion strength
- Link distance
- Collision radius
- Python 3.11+
- NetworkX 3.0+
- Modern web browser with JavaScript enabled
We welcome contributions! Whether it's:
- π Bug reports
- π‘ Feature suggestions
- π Documentation improvements
- π§ Code contributions
Please see our Contributing Guide for details on:
- Development setup
- Coding standards
- Pull request process
- Testing guidelines
# Fork and clone the repository
git clone https://github.com/YOUR_USERNAME/CVEMaps.git
cd CVEMaps
# Install in development mode
pip install -e ".[dev]"
# Run tests
pytest
# Format code
black cvemaps/
ruff check cvemaps/This project is licensed under the MIT License - see the LICENSE file for details.
- CVE Project for providing comprehensive CVE data
- D3.js for powerful visualization capabilities
- NetworkX for graph processing and algorithms
- Tailwind CSS for modern, responsive styling
- Documentation: See QUICKSTART.md for quick setup guide
- Changelog: See CHANGELOG.md for version history
- Issues: Report bugs or request features on GitHub Issues
- Discussions: Join conversations on GitHub Discussions
- CVEs Analyzed: 315,567+
- CNAs Tracked: 400+
- CWEs Mapped: 800+
- Visualizations: 10 interactive types
- Update Frequency: Every 3 hours
- Data Window: Last 365 days
For questions or feedback, please open an issue on GitHub.
Last Updated: Auto-generated by GitHub Actions