Open-source licences
Last updated 2026-09-22
The tools on this site run on open-source software that is downloaded to your browser and runs there. This page lists every third-party engine, model, data set and font the site ships, the version in use, its licence and where its source is published.
Two components carry copyleft licences. The video and audio engine is FFmpeg compiled to WebAssembly, licensed GPL-2.0-or-later; its licence text and a notice with the exact package version and the source links are served next to it in /ffmpeg/. The HEIC decoder libheif is LGPL-3.0; its notice and licence texts are in /third-party/libheif-js/. Everything else is under a permissive licence.
| Component | Used for | Version | Licence | Source and notice |
|---|---|---|---|---|
| FFmpeg (ffmpeg.wasm core) | Video and audio compression, conversion, trimming and captions | @ffmpeg/core 0.12.10 | GPL-2.0-or-later | Source · Notice and licence |
| ffmpeg.wasm JavaScript API | Running the FFmpeg core in a worker | @ffmpeg/ffmpeg 0.12.15, @ffmpeg/util 0.12.2 | MIT | Source |
| libheif (libheif-js) | Opening HEIC and HEIF photos from iPhones | libheif-js 1.23.2 | LGPL-3.0 | Source · Notice and licences |
| Lucide icons | The tool and section icons (SVG shapes copied into the site at build time) | lucide-react 1.47.0 | ISC | Source |
| PDF.js | Reading, previewing and rendering PDFs | pdfjs-dist 6.3.289 | Apache-2.0 | Source |
| pdf-lib and fontkit | Creating and editing PDFs | pdf-lib 1.17.1, @pdf-lib/fontkit 1.1.1 | MIT | Source |
| qpdf (qpdf-wasm) | Protecting, unlocking and repairing PDFs | @neslinesli93/qpdf-wasm 0.3.0 (qpdf 12.2.0) | Apache-2.0 (qpdf), ISC (wrapper) | Source · Notice |
| Tesseract OCR (tesseract.js) | OCR PDF and image to text | tesseract.js 7.0.0, tesseract.js-core 7.0.0 | Apache-2.0 | Source · Notice |
| Tesseract trained data (English, Arabic) | OCR language models | @tesseract.js-data/eng and /ara 1.0.0 (4.0.0_best_int) | Apache-2.0 (models), MIT (packaging) | Source · Licence |
| ONNX Runtime Web | Running the background-removal, diacritisation and speech models | onnxruntime-web 1.30.0 | MIT | Source · Notice |
| U²-Net (u2netp model) | Background remover | u2netp | Apache-2.0 (model), MIT (rembg export) | Source · Notice and licences |
| Whisper speech-recognition models (sherpa-onnx export) | Video to transcript, audio to text, subtitles | Whisper tiny, base and small | MIT (Whisper), Apache-2.0 (sherpa-onnx) | Source · Notice and licences |
| libtashkeel model | Arabic diacritizer (tashkeel) | libtashkeel ONNX model | MIT | Source · Notice and licences |
| mammoth | Reading Word documents (Word to PDF) | 1.12.3 | BSD-2-Clause | Source |
| docx | Writing Word documents (PDF to Word, CV, transcripts) | 9.7.1 | MIT | Source |
| exifr | Reading photo orientation and metadata | 7.1.3 | MIT | Source |
| jsQR | QR code scanner | 1.4.0 | Apache-2.0 | Source |
| node-qrcode | QR code generators | qrcode 1.5.4 | MIT | Source |
| JsBarcode | Barcode generator | 3.12.3 | MIT | Source |
| USDA SR28 food composition data (FoodData Central SR Legacy) | Food calorie calculator (nutrition values and portion weights) | SR28, September 2015 (distributed as SR Legacy, April 2018) | Public domain (US Government work); FoodData Central data CC0 1.0 | Source |
| CLIP ViT-B/32 image model (int8 ONNX) | Food calorie calculator (recognising food in a photo, on the device) | OpenAI CLIP ViT-B/32 image tower, re-stored as int8 | MIT (OpenAI CLIP weights and OpenCLIP export) | Source · Notice and licence |
| Carlito font | CV builder and document PDFs | @expo-google-fonts/carlito 0.4.1 | OFL-1.1 | Source · Licence |
| Cairo font | Site headings and Arabic text | @fontsource-variable/cairo 5.1.1 | OFL-1.1 | Source · Licence |
| Inter font | Site body text | @fontsource-variable/inter 5.1.1 | OFL-1.1 | Source · Licence |
| IBM Plex Sans Arabic font | Transcript PDFs and burned-in captions | @expo-google-fonts/ibm-plex-sans-arabic 0.4.2 | OFL-1.1 | Source · Licence |
| Next.js and React | The site itself | next 16.3.5, react 19.2.8 | MIT | Source |
| GPT-2 small language model (int8, ONNX Model Zoo export) | AI text detector (language-model signals, English) | GPT-2 124M, gpt2-lm-head-10.onnx re-stored as int8 | Modified MIT (GPT-2 weights), Apache-2.0 (ONNX Model Zoo), MIT (js-tiktoken vocabulary packaging) | Source · Notice and licences |