# leigest519/ScreenCoder

ScreenCoder — Turn any UI screenshot into clean, editable HTML/CSS with full control. Fast, accurate, and easy to customize.

Repository: https://github.com/leigest519/ScreenCoder
Canonical: https://ross.abutalabs.com/products/screencoder
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-24T05:23:53+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 35, longevity 28
- inputs: {"age_days": 401, "days_push": 9, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2951, forks 312 (observed 2026-08-28T04:07:31.879906+00:00)

## What it is
ScreenCoder is a UI-to-code generation system that converts screenshots or design mockups into clean, editable HTML/CSS using a modular multi-agent architecture combining visual understanding, layout planning, and code synthesis. It also supports customized layout and styling modifications and includes ScreenBench, a benchmark of 1000 real-world web screenshots with corresponding HTML.

## Use cases
- convert a UI screenshot into HTML/CSS code
- turn a design mockup into editable front-end code
- generate web page code from an image
- prototype interfaces quickly from screenshots
- benchmark visual-to-code generation models
- customize generated layout and styling from a design draft

## When to choose
- you need to translate screenshots or mockups into production-ready HTML/CSS automatically
- you want editable, customizable generated front-end code rather than static images
- you need a benchmark for evaluating visual-to-code generation systems
- you want a modular multi-agent pipeline for UI understanding and code synthesis

## When to avoid
- you need pixel-perfect hand-crafted production code without AI generation
- you work with non-web targets like native mobile or desktop UIs
- you cannot run GPU-backed multimodal models or lack API access for inference

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, llm-inference, agent-framework, image-processing
- domain: artificial-intelligence, large-language-models, web-development, frontend, computer-vision
- platform: python
- tags: ui-to-code, screenshot-to-html, multimodal-agents, design-to-code, front-end-automation, html-css-generation, vision-language-models, benchmark, code-generation, web-server, docker

## Member repositories
- leigest519/ScreenCoder (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:31.879906+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:32:39.804255+00:00, confidence not recorded.
  - readme: https://github.com/leigest519/ScreenCoder (fetched 2026-08-28T04:07:31.879906+00:00, sha f11796134fa6)
- Data as of 2026-08-30T08:39:29.467469+00:00.