45645678a

45645678a / scholar-mcp

Public

Local paper tool MCP server for AI IDEs - search, download (Sci-Hub/arXiv/Unpaywall), AI analysis, code-based recommendation, citation graph

77
7
69% credibility
Found Mar 25, 2026 at 77 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A tool that integrates with AI coding assistants to search academic papers from multiple sources, download PDFs prioritizing open access, perform AI analysis, recommend based on code, and visualize citation graphs.

How It Works

1
📚 Discover the paper helper

You hear about a handy tool that finds research papers, downloads them, and even analyzes them right inside your coding app with AI.

2
⬇️ Set it up quickly

With one easy command, you download and prepare the tool so it's ready to use.

3
🔗 Connect to your AI assistant

You link it to your favorite coding app's smart helper, and everything is set.

4
💬 Chat about papers

In your app's chat, you simply ask to search for papers on your topic, like 'transformer models', and get top results instantly.

5
📥 Grab and review papers

Tell it to download a paper or analyze its key ideas, and you get the full file plus smart summaries.

6
🔍 Get smart recommendations

It scans your code project and suggests related papers, or draws a map of citations to see connections.

🎉 Research supercharged

Now your work is boosted with fresh papers, insights, and visuals, making you more productive and informed.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 77 to 77 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 scholar-mcp?

Scholar-mcp is a Python MCP server that plugs into AI IDEs like Cursor, VS Code Copilot, and Claude Code, letting you search academic papers across nine free sources, download PDFs via Unpaywall, arXiv, or Sci-Hub, run AI analysis on full text, recommend papers based on your codebase, and generate citation graphs. It solves the hassle of tabbing out of your IDE for research—query "transformer models," grab DOIs like 10.48550/arXiv.2005.14165, or analyze imports for relevant reads, all locally without cloud dependencies. Install via pip, set env vars for your OpenAI-compatible API, and restart your IDE.

Why is it gaining traction?

Unlike browser-based tools or clunky CLI scrapers, it embeds directly into your workflow as a local GitHub Copilot alternative for scholars, with concurrent 9-source searches ranked by relevance, citations, and recency, plus batch downloads and Mermaid citation viz. Developers dig the code-scanning recommender that spots PyTorch or scikit-learn usage and suggests papers, and the flexible AI analysis using any LLM endpoint. No API keys needed for core search/download beats paid services.

Who should use this?

ML engineers building prototypes who scan repos for related papers, PhD students in Cursor drafting theses with citation graphs, or researchers juggling local paper news and local papers near me via IDE chats. Perfect for cpp mcphee scholar workflows or john mcphee scholar deep dives, as a local GitHub actions runner for academic refs without leaving VS Code.

Verdict

Grab it if you're in AI/research IDEs—solid features, 40 tests, clear docs, and MIT license make it usable now despite 66 stars and 0.7% credibility score signaling early days. Mature enough for daily drivers, but watch for Sci-Hub legal notes in institutional setups.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.