deepxiv

Retrieve open-access academic papers by topic or ID with progressive reading depths.

2|1|Updated Apr 19, 2026

Search Tags:#literature review #semantic scholar #arxiv #paper search #progressive reading #trending research #section retrieval

Ready to run

npx skills add https://github.com/raja21068/AutoResearch --skill deepxiv-raja21068

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: deepxiv
Source: https://github.com/raja21068/AutoResearch/tree/main/skills/aris/deepxiv
Command: npx skills add https://github.com/raja21068/AutoResearch --skill deepxiv-raja21068

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

DeepXiv search and progressive reading removes the time sink of loading full papers too early by letting you quickly find relevant work and then escalate depth only when needed.

Core Features & Use Cases

Layered discovery and reading: browse trending papers or search by topic/ID, then fetch a brief, a head/section map, or a single named section.
Token-efficient workflow: escalate from search → brief → head → section to avoid unnecessary full-paper reads.
Fallback-aware retrieval: prefers an ARIS adapter script when available, otherwise falls back to the DeepXiv CLI.

Use case: you need background on a specific method and want to gather enough evidence for your literature review without downloading and reading entire PDFs.

Quick Start

Ask for: /deepxiv "2409.05591" - section: Introduction to retrieve only the Introduction section from that paper.

Frequently Asked Questions about deepxiv

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

FAQPage Schema

How do I search arXiv papers and read specific sections without loading the full PDF?▼

You can search arXiv and Semantic Scholar metadata by topic or paper ID, then progressively retrieve a brief, head, or named section to avoid loading full PDFs unnecessarily. Escalate depth from search to brief to section only when needed.

What is progressive reading for literature reviews and how does it save tokens?▼

Progressive reading lets you fetch academic papers at brief, head, or section depth, minimizing token usage by escalating from search to brief to section. This avoids loading full open-access papers until you confirm relevance.

How can I monitor trending research papers across arXiv and Semantic Scholar?▼

You can browse trending papers and search across arXiv, DeepXiv, and Semantic Scholar metadata workflows. Use layered discovery to fetch briefs first, then escalate to head or section depth for targeted evidence gathering.

Do I need an ARIS adapter script to retrieve open-access academic papers?▼

An ARIS adapter script is preferred for DeepXiv-compatible retrieval but is not strictly required. The system uses fallback-aware retrieval, falling back to the DeepXiv CLI when the adapter is unavailable.

Can I retrieve just the Introduction section of a paper using its arXiv ID?▼

Yes, you can retrieve a specific named section like the Introduction by providing the paper ID and specifying the section depth. This targeted section retrieval avoids loading the entire paper when you only need specific evidence.

What are the limitations of using progressive reading for academic paper search?▼

Progressive reading is limited to open-access papers from arXiv, DeepXiv, and Semantic Scholar. It enforces depth-escalation to minimize full-paper loading, meaning you cannot bypass brief or head levels to jump directly to full content.