A proof of concept for downloading from Comix.to.
Latest updates to Comix.to added Pro obfuscator.io to the JavaScript, which contains the rotated decryption function as well as the URL signing function.
This PoC automatically patches the obfuscated secure-[hash].js file, and uses Node.js to emulate the decryption and signing functions.
The Go version of the script can download and unscramble images, but it requires the dependencies in order to run a headless browser.
go run rod_downloader.go "https://comix.to/title/2vzj2-im-in-love-with-the-villainess"Python version of the script DOES NOT unscramble the images.
Install dependencies with uv sync, then download a chapter:
uv run python main.py "https://comix.to/title/2vzj2-im-in-love-with-the-villainess/8998931-chapter-61"Download the first three English chapters from a manga id using eight page-download threads:
uv run python main.py 2vzj2 --language en --limit 3 --workers 8Downloaded pages are written under downloads/ by default. Existing non-empty page files are skipped on repeat runs, and a .download-complete.json marker is written after a chapter finishes successfully so future runs can skip the whole chapter. Use --dry-run to inspect selected chapters without downloading images, and --overwrite to ignore existing pages and completion markers.
A partially deobfuscated version of the secure-[hash].js file is included in this repo for reference. See secure_dec.js.
By default, the script skips chapters with duplicate indices, (e.g. only one translation of Chapter 1 will be downloaded).
However, the source sometimes label different translations of the same chapter with different indices (e.g. 1 and 1.5). In these cases, the script will download both chapters by default.