Logo Lanfrica

gabrielmahia/africa-leapfrog-kit

Domaine:

geospatialagricultureenvironment and energy

Type de record:

software
Créateur:
gab
Hôte:
AI-powered infrastructure intelligence toolkit for Kenya and East Africa. Geospatial gap analysis, NDVI crop stress, county data dashboards. # Africa Leapfrog Toolkit 🌍 > **Decision infrastructure for East Africa** — AI-powered gap analysis, NDVI crop stress > monitoring, county early warning, and chama/SACCO trust scoring. ## What This Is Africa doesn't need to wait for full institutional maturity to get institutional-grade decision support. This toolkit approximates the functions of expensive planning systems using open data, geospatial analysis, and AI — making them available in under 2 minutes for any county, ward, or community in Kenya and East Africa. **Built on patterns from:** - *Learning Geospatial Analysis with Python 3rd Ed.* (Packt) — Ch.1 SimpleGIS, Ch.8 NDVI - *Python Data Analysis 3rd Ed.* (Packt) — group statistics and trend analysis - *ML for Algorithmic Trading 2nd Ed.* (Packt/Stefan Jansen) — Kelly Criterion sizing ## Modules | Module | Function | |--------|----------| | `gap_mapper.py` | Infrastructure gap scoring (water, health, education, energy) across 20+ Kenya counties | | `ndvi_monitor.py` | Crop stress / vegetation health — NDVI pattern, open satellite data | | `early_warning.py` | County food security brief — drought phase, rainfall, NDVI, price anomaly | | `chama_trust.py` | SACCO/chama group health scoring + Kelly Criterion lending capacity | | `county_dashboard.py` | County statistics (coming soon) | | `procurement_watch.py` | Tender anomaly detection (coming soon) | ## MCP Integration This toolkit connects to the full coordination infrastructure stack: - **wapimaji-mcp** → drought/NDMA data - **afya-mcp** → health facility gaps - **elimu-mcp** → school access - **ardhi-mcp** → land/infrastructure - **soko-mcp** → market access ## Running Locally ```bash git clone github.com cd africa-leapfrog-kit pip install -r requirements.txt streamlit run app.py ``` ## Data Sources (Production) | Data | Source | Cost | |------|--------|------| | NDVI (250m, 16-day) | NASA MODIS TERRA | Free — earthdata.nasa.gov | | Rainfall anomaly …