sammwyy

sammwyy / cosmisum

Public

A Python pipeline for analyzing manga, comics, and documents. Extracts panels from PDFs, performs OCR, and uses LLM to generate summaries, tags, and genre classification.

31
4
100% credibility
Found Feb 08, 2026 at 24 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

Cosmisum is a Python pipeline for analyzing manga, comics, and documents by extracting panels from PDFs, performing OCR, and using LLMs to generate summaries, tags, and genre classifications.

Star Growth

See how this repo grew from 24 to 31 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is cosmisum?

Cosmisum is a Python pipeline that processes PDF comics, manga, or documents, extracting panels, running OCR for text, and using an OpenAI-compatible LLM to generate plot summaries, thematic tags, and genre classification. Run it via a simple CLI like `python cosmisum.py input.pdf` after a quick python github install with pip and prerequisites like Poppler and Tesseract. It solves the hassle of manually dissecting visual stories for analysis or archiving, delivering console output or saved files in markdown or JSON.

Why is it gaining traction?

This stands out as a lean python pipeline example on GitHub, blending PDF handling, OCR, and LLM chunking into a cross-platform tool that supports local models via custom API endpoints. Developers dig the uniform token distribution for efficient analysis without wasting LLM credits, plus easy tweaks for multi-language OCR. It's a practical python pipeline framework for document classification tasks, trending in python github projects for its no-fuss setup over bloated alternatives.

Who should use this?

Manga collectors batch-analyzing chapter PDFs for tags and genres, researchers classifying comic archives, or indie devs prototyping python package pipelines for visual content. Ideal for AI hobbyists experimenting with python github copilot workflows on scanned docs, or content curators needing quick summaries without full reads.

Verdict

Grab it for a weekend hack on comic analysis—solid docs and MIT license make it a credible python github module starter, despite 27 stars and 1.0% credibility score signaling early maturity. Test on your PDFs first; scale up once you add custom pipelines.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.