deepxiv

Search and progressively read open-access academic papers through the DeepXiv SDK.

Updated Jul 9, 2026
One-click install
npx skills add https://github.com/Lingjie-wang/autoRL --skill deepxiv-lingjie-wang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deepxiv
Source: https://github.com/Lingjie-wang/autoRL/tree/main/Auto-claude-code-research-in-sleep/skills/deepxiv
Command: npx skills add https://github.com/Lingjie-wang/autoRL --skill deepxiv-lingjie-wang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires deepxiv-sdk.

What problem does it solve? Reading full academic papers consumes excessive context and time when you only need a summary or one section. This Skill provides layered, section-level access to open-access papers via DeepXiv so you only load the depth you need. ## Core Features & Use Cases - Progressive Reading: Escalate from search to brief, section map, single section, and full paper only when necessary. - Multiple Query Modes: Search by topic, arXiv ID, Semantic Scholar ID, trending papers, or DeepXiv web search. - Research Wiki Integration: Automatically ingests depth-read papers into a project research-wiki when one exists. - Use Case: Ask for a brief on arXiv paper 2409.05591, then read only its Introduction section, avoiding a full-paper load. ## Quick Start Use the deepxiv skill to search for recent papers on agent memory and show the top five results.

Frequently Asked Questions about deepxiv

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I read only one section of an arXiv paper?

Use the deepxiv skill with the paper's arXiv ID and the section override, for example "/deepxiv 2409.05591 - section: Introduction". This fetches only that named section instead of the full paper, saving context tokens.

What is the difference between DeepXiv, arXiv, and Semantic Scholar skills?

DeepXiv provides progressive section-level reading of papers, the arXiv skill handles batch search and PDF download via the arXiv API, and Semantic Scholar supplies published venue metadata and citation counts. They are complementary sources, not replacements.

How do I install the DeepXiv CLI?

Install it with pip using the command "pip install deepxiv-sdk". On first use, the deepxiv CLI auto-registers a free token and stores it in your ~/.env file.

What happens if the deepxiv_fetch.py adapter is not found?

The skill falls back to raw deepxiv CLI commands documented for each operation, such as "deepxiv search" or "deepxiv paper --brief". If the CLI itself is missing, you are given the install command and pointed to alternatives like the arxiv skill.

Does deepxiv update a research wiki automatically?

Yes, but only when a research-wiki/ directory exists in the project. Papers read in depth (brief, head, section, or full) are ingested via the research_wiki.py helper; plain search hits without a depth read are skipped.