explaingit

xuezheng627/research-radar-paper-vault

★ 22PythonAudience · researcherComplexity · 3/5Setup · moderate

TLDR

Two Codex AI skills that together let researchers get a daily email digest of new academic papers matching their keywords, then save the ones they read into a local personal paper library website.

Mindmap

mindmap
  root((research-radar-paper-vault))
    Skill 1 daily-literature-digest
      Search Crossref and OpenAlex and arXiv
      Email daily summary
      Local Markdown archive
    Skill 2 paper-vault
      Local static website library
      Cards organized by category
      Unread queue
    Tech stack
      Python
      Codex AI tool
      Gmail
      HTML and CSS and JS
    Setup
      Copy skill folders
      Configure keywords and email
      Run Python local server
Click or tap to explore — scroll the page freely

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

Things people build with this

USE CASE 1

Get a daily email digest of new papers on your research topic pulled from arXiv, Crossref, and OpenAlex without visiting each database manually.

USE CASE 2

Build a local website that displays all the academic papers you have fully read, organized by category.

USE CASE 3

Keep a queue of papers you found interesting but have not yet read the full text of, separate from your main library.

USE CASE 4

Archive your daily paper digest as Markdown files on your computer for later offline reference.

Tech stack

PythonHTMLCSSJavaScriptCodexGmail

Getting it running

Difficulty · moderate Time to first run · 30min

Requires the Codex AI tool, a Gmail account connected to Codex, and the computer to be awake at the scheduled daily digest time.

No license information was mentioned in the explanation.

In plain English

This repository contains two skills for use with Codex, an AI coding tool made by OpenAI. The two skills are designed to work together as a reading pipeline for researchers who want to track new academic papers in their field without manually checking multiple databases each day. The first skill, daily-literature-digest, searches Crossref, OpenAlex, and arXiv each day using keywords you define, then emails a summary of recent papers to an address you specify through a Gmail connection in Codex. You configure your email address, keywords, preferred language, and timezone, and Codex sets up a daily automation to send the digest at a scheduled time. A local Markdown archive of results is saved alongside the emails. The automation depends on the Codex runner and your computer being awake at the scheduled time. The second skill, paper-vault, builds a local static website that serves as a personal paper library. After reading the full text of a paper from the digest, you ask Codex to add it to the vault, where it appears as a card organized by category. Papers you have not yet obtained full-text access to go into a separate queue rather than the main view, keeping the library limited to papers you have actually read. The vault is a folder of HTML, CSS, and JavaScript files you serve locally with a simple Python command. Both skills are installed by copying the skill folders into a Codex skills directory and restarting Codex. The README includes both English and Chinese instructions. Personal generated files, such as digest archives, state data, PDFs, and login traces, are explicitly excluded from what should be committed to the repository.

Copy-paste prompts

Prompt 1
I want to set up the daily-literature-digest Codex skill to email me new machine learning papers every morning. Walk me through installing the skill, configuring my Gmail connection, setting keywords, and scheduling the daily run.
Prompt 2
Show me how to add a paper to my paper-vault after reading it: what command do I give Codex and how does the card appear on the local website?
Prompt 3
How do I run the paper-vault website locally? Show the Python command to serve the static HTML files and open it in a browser.
Prompt 4
What happens to papers I find in the digest but cannot access the full text of? How does the paper-vault skill handle them differently from papers I have read?
Open on GitHub → Explain another repo

← xuezheng627 on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.