The Face Analysis developer docs
Three free, open-source building blocks from The Face Analysis: an MCP server that lets AI assistants look up sourced face measurement statistics, the dataset behind those statistics as JSON and CSV, and an embeddable widget that puts the face shape finder on any website.
Start here
| I want to… | Go to |
|---|---|
| Let Claude, ChatGPT, Codex or Cursor look up face measurement data | MCP server |
| Download the numbers as JSON or CSV | Dataset |
| Put a face shape finder on my website | Embed widget |
| Try everything in two minutes | Quickstart |
What is published
| Package | Version | What it is | Source |
|---|---|---|---|
thefaceanalysis-mcp | 0.1.0 | Read-only MCP server, stdio and hosted HTTP | GitHub |
thefaceanalysis-data | 0.1.0 | Statistics, type rules and thresholds as JSON and CSV | GitHub |
thefaceanalysis-embed | 0.1.0 | Web component and React wrapper for the widget | GitHub |
What it is based on
All numbers come from 102 adult faces (49 women, 53 men) of the Face Research Lab London Set, measured with a 478-point MediaPipe landmarker. They describe those faces, not a population, and the face type splits show how the detectors divide them, not how common a type is. The methodology states every threshold and what a photo cannot measure.
Licenses
- Code (MCP server, dataset package, widget, these docs): MIT.
- Aggregate numbers (JSON and CSV): CC BY 4.0. Cite The Face Analysis, link the statistics page and credit the photo set.