git404hub

what is makeup-match-prototype fr?

fanshouheng/makeup-match-prototype — explained in plain English

Analysis updated 2026-05-18

78TypeScriptAudience · generalComplexity · 3/5LicenseSetup · moderate

tl;dr

A privacy focused web app that analyzes a user's face proportions entirely in their browser to recommend makeup tutorial creators with a similar face shape.

vibe map

mindmap
  root((makeup match))
    What it does
      Local face proportion analysis
      Creator matching
      No login needed
    Tech stack
      TypeScript
      MediaPipe
      React
    Privacy
      Browser only processing
      No photo upload
      Session discarded on close
    Use cases
      Find similar creators
      View tutorials
      Give match feedback
      Creator library application

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

what do people make with this?

VIBE 1

Upload a front facing photo to find makeup tutorial creators whose face shape is similar to your own.

VIBE 2

Browse a matched creator's profile and featured tutorial for makeup style ideas that suit your face.

VIBE 3

Give quick feedback on whether a suggested creator match feels accurate, without needing to log in.

VIBE 4

Apply as a creator to be added to the public matching library with an authorized profile photo.

what's the stack?

TypeScriptMediaPipeReact

how it stacks up fr

fanshouheng/makeup-match-prototypeadrienckr/notslopalchemz/solana-pumpfun-token-bundler
Stars787878
LanguageTypeScriptTypeScriptTypeScript
Setup difficultymoderateeasyhard
Complexity3/52/54/5
Audiencegeneralwriterdeveloper

Figures from each repo's GitHub metadata at analysis time.

how do i run it?

Difficulty · moderate time til it works · 30min

Uses MediaPipe face landmark detection in the browser, the public creator library and review process are hosted separately from the open source code.

GNU AGPL v3.0, if you run a modified version as a network service, you must share your source code with its users.

in plain english

Makeup Match Prototype, also called LOOK AI, is a privacy focused tool that helps someone find makeup tutorial creators whose face shape is similar to their own. A user uploads a clear front facing photo, and the app compares facial proportions entirely inside the browser on the user's own device, then shows creators whose face structure is a closer match, along with links to their profiles and featured tutorials. It does not try to identify who someone is and does not judge appearance. To use it, someone picks or takes a clear, evenly lit, front facing photo with no obstructions. The app checks photo quality and analyzes facial proportions, then shows matching creators. No login is required, and the user can mark a result as a good or poor match, or share it. From there, the user can visit a creator's profile or featured tutorial to find a makeup style that suits them. Photos with a neutral expression, closed lips, and a straight on head position give more stable results, and only one clear face should appear in the photo. Privacy is central to the design. The original uploaded photo is never sent to or stored on a server, the facial proportions extracted from it are never uploaded, and the similarity comparison happens entirely in the browser. Closing or refreshing the page discards the photo and analysis. The only data the app fetches from a server is the public library of creators who have already been reviewed and approved. Creators can apply to be added to that library by submitting their name, social media profile, contact email, an authorized clear photo of themselves, and an optional featured tutorial. Submitted applications are not made public automatically, the team verifies profile ownership and photo authorization before adding a creator to the public library, and this is a separate process from the private, on device matching used by ordinary visitors. The project is licensed under GNU AGPL v3.0, which means that if someone deploys a modified version of the app over a network, they must also make the corresponding source code available to users. The open source license only covers the code itself, not the production creator database, real photos, facial feature data, or contact and authorization records.

prompts (copy fr)

prompt 1
Explain how makeup-match-prototype keeps a user's uploaded photo from ever leaving their browser.
prompt 2
Walk me through how a creator applies to be added to the public matching library in this project.
prompt 3
What does the AGPL v3.0 license require if I deploy a modified version of this app on my own server?
prompt 4
Describe the steps a user follows from uploading a photo to finding a matched makeup tutorial.

Frequently asked questions

what is makeup-match-prototype fr?

A privacy focused web app that analyzes a user's face proportions entirely in their browser to recommend makeup tutorial creators with a similar face shape.

What language is makeup-match-prototype written in?

Mainly TypeScript. The stack also includes TypeScript, MediaPipe, React.

What license does makeup-match-prototype use?

GNU AGPL v3.0, if you run a modified version as a network service, you must share your source code with its users.

How hard is makeup-match-prototype to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is makeup-match-prototype for?

Mainly general.

peek the repo → explain another one

This repo across BitVibe Labs

double-check against the repo, no cap.