gitwtfhub

wtf is covid-sanity?

karpathy/covid-sanity — explained in plain English

Analysis updated 2026-07-17 · repo last pushed 2020-05-03

★ 393PythonAudience · researcherComplexity · 3/5DormantSetup · moderate

TL;DR

A searchable database of COVID-19 research papers that uses machine learning to find similar papers, helping researchers cut through the flood of preprint publications.

Mindmap

mindmap
  root((repo))
    What it does
      Collects preprint papers
      Finds similar abstracts
      Searchable database
      Shows Twitter discussion
    Tech stack
      Python
      Flask
      Machine learning
    Use cases
      Research discovery
      Public health triage
      Journalism research
    Audience
      Researchers
      Public health officials
      Medical students

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

Why would anyone build with this?

REASON 1

Search and sort COVID-19 preprint papers from bioRxiv and medRxiv by topic.

REASON 2

Find papers similar to one you're reading using abstract-based machine-learning matching.

REASON 3

Deploy your own instance to track a fast-moving body of research during a crisis.

REASON 4

See Twitter discussion linked to a specific research paper.

What's in the stack?

PythonFlask

How it stacks up

karpathy/covid-sanityfacebookresearch/spdljmmy9609-design/gpt-pp
Stars393393396
LanguagePythonPythonPython
Last pushed2020-05-032026-06-25—
MaintenanceDormantActive—
Setup difficultymoderatemoderatemoderate
Complexity3/53/54/5
Audienceresearcherdeveloperops devops

Figures from each repo's GitHub metadata at analysis time.

How do you spin it up?

Difficulty · moderate Time to first run · 30min

Needs an hourly update script and Flask hosting to keep the paper database current.

Yoink these prompts

Prompt 1
Help me set up covid-sanity locally using Flask so I can browse COVID-19 preprint papers.
Prompt 2
Show me how this repo's abstract-similarity search works so I can adapt it for a different research topic.
Prompt 3
Write the hourly update script logic this repo uses to keep the paper database fresh.
Prompt 4
Explain how I could deploy this Flask app to a cheap host like Linode, based on this repo's setup.
Prompt 5
Help me add the optional Twitter discussion feature described in this repo to my own paper-tracking tool.

Frequently asked questions

wtf is covid-sanity?

A searchable database of COVID-19 research papers that uses machine learning to find similar papers, helping researchers cut through the flood of preprint publications.

What language is covid-sanity written in?

Mainly Python. The stack also includes Python, Flask.

Is covid-sanity actively maintained?

Dormant — no commits in 2+ years (last push 2020-05-03).

How hard is covid-sanity to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is covid-sanity for?

Mainly researcher.

View the repo → Decode another repo

This repo across BitVibe Labs

Don't trust strangers blindly. Verify against the repo.

Other repos in this lane