Wind site intelligence

Wind site suitability, computed.

Six-factor scoring. Bias-corrected against ERA5 and CERRA reanalysis. ISO 9613-2 noise modelling. Open source, free APIs, callable from Claude Desktop and Cursor.

Run an analysisView on GitHub
Tests passing
0vitest
Scoring factors
0weighted
ERA5 · CERRA
0reanalysis sources
Noise model
ISO 9613-2standard
What it measures

Three layers of evidence per site.

WIND
Resource and stability

Long-term mean speed, Weibull distribution, directional consistency. NASA POWER raw, ERA5 and CERRA bias-corrected at 50m and extrapolated to hub height.

TERRAIN
Slope, roughness, flow

Open-Elevation samples, slope and aspect derived locally, surface roughness class informs the wind shear exponent for hub-height extrapolation.

CONSTRAINTS
Grid, land use, planning

Distance to transmission lines and substations, protected zones and residential buffers, planning precedent from existing wind installations within 20km.

Model Context Protocol

Call WindForge from your editor.

Six tools served over MCP stdio. Add the package to Claude Desktop, Cursor or any MCP-compatible client and ask in plain language.

Six tools exposed: analyse_site assess_site_polygon calculate_aep list_turbines fetch_wind_history detect_constraints

Read the MCP setup guide →
$ npx -y @jamieblair/windforge-mcp