rkka02

rkka02 / Annot

Public

PDF Research Assistant — Read papers with AI

19
0
100% credibility
Found Mar 30, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

Annot is a local app for organizing PDF research papers into folders, reading and annotating them, and chatting contextually with AI using your machine's existing AI tool logins.

How It Works

1
🔍 Discover Annot

You stumble upon this handy tool for reading research papers with a smart chat helper right by your side.

2
🚀 Start Annot on your computer

You get the app running locally with one easy launch, and your personal research space opens up.

3
🧠 Link your AI thinking partner

You connect your existing AI login in settings, test it works, and now it can help with your papers.

4
📁 Build your paper library

You create folders in the side explorer and upload your PDF research papers to stay organized.

5
📖 Read and mark up a paper

You open a PDF, scroll smoothly, select text to highlight important bits, and erase if needed.

6
💬 Ask the AI anything

You type questions in the chat panel next to the paper, and get clever summaries or explanations instantly.

🎉 Unlock research superpowers

Your highlights save right into the PDFs, chats remember everything, and you breeze through papers smarter than ever.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 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 Annot?

Annot is a local-first PDF research workspace built in JavaScript with Next.js and React, letting you organize papers in folders, read them in a built-in viewer, add annotations that save directly to the PDF files, and chat with AI about specific documents or folders. It hooks into your existing Codex or Claude Code CLI logins—no API keys needed—and supports streaming responses with math rendering via KaTeX. Perfect for the read-highlight-ask loop without leaving your filesystem-backed Annot folder.

Why is it gaining traction?

Unlike generic PDF github readers or AI chatbots, Annot ties annotations, highlights (including eraser mode), and context-aware chats to your actual PDFs and folders, writing native annotations back via PyMuPDF for persistence across tools. The resizable chat panel, vertical scroll viewer, and provider switching (Codex/Claude) make it snappier than browser-based pdf github ai editors, with no cloud dependency. Devs dig the seamless workflow for pdf github tools like annotating, annotieren, or building annotopia-style setups.

Who should use this?

Academic researchers skimming 10+ papers weekly, ML engineers doing lit reviews on arXiv dumps, or PhD students needing quick summaries and equation breakdowns tied to highlights. Ideal if you're already running Codex/Claude locally and want a pdf github editor that handles annotations, annotationen, and folder sessions without export hassles.

Verdict

Worth a spin for pdf github ai workflows if you have the prereqs (Node 20+, PyMuPDF, poppler)—it's a solid prototype at 19 stars, but the 1.0% credibility score flags early-stage risks like sparse tests. Fork and tweak for production; don't bet the farm yet.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.