deepxiv

Search and incrementally retrieve academic paper content with staged output.

Updated Apr 21, 2026
One-click install
npx skills add https://github.com/Shallow-W/llm-wiki --skill deepxiv-shallow-w
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deepxiv
Source: https://github.com/Shallow-W/llm-wiki/tree/main/.claude/skills/deepxiv
Command: npx skills add https://github.com/Shallow-W/llm-wiki --skill deepxiv-shallow-w

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you find relevant academic papers and read only the right amount at the right time, reducing token waste and time spent downloading or scanning full papers.

Core Features & Use Cases

  • Layered paper access: search → brief → head → single section so you can match reading depth to your question.
  • Special retrieval modes: trending papers over a time window and DeepXiv web search for broader discovery.
  • Metadata lookup: Semantic Scholar integration by ID to enrich results when available.
  • Optional wiki ingestion: if a research-wiki directory is present, it can ingest meaningfully read papers (brief/head/section/full) to keep your research notes up to date.

Quick Start

Use the deepxiv skill to fetch a brief summary for an arXiv paper by asking: deepxiv 2409.05591 - brief.

Frequently Asked Questions about deepxiv

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

FAQPage Schema
How do I read arXiv papers progressively without downloading the full text?

You can read arXiv papers progressively by searching and retrieving content in stages: search, brief, head, and single section. This layered approach matches reading depth to your specific question, reducing token waste and time spent downloading full papers.

What is the best way to discover trending academic papers over a specific time window?

Discover trending academic papers by using the trending research retrieval mode. It allows you to configure a specific time window to find popular research, providing JSON-formatted fallbacks to ensure you get metadata even if primary sources are unavailable.

Can I retrieve academic paper metadata using Semantic Scholar IDs?

Yes, you can retrieve academic paper metadata by performing a Semantic Scholar integration using a specific paper ID. This enriches your search results with detailed metadata when available, helping you evaluate papers before reading.

Does DeepXiv support saving research notes from arXiv searches to a local wiki?

Yes, DeepXiv supports optional wiki ingestion. If a research-wiki directory is present, it can ingest meaningfully read papers (brief, head, section, or full) to keep your research notes updated automatically.

How do I search arXiv by topic or specific paper ID to get a brief summary?

You can search arXiv by topic or specific paper ID to get a brief summary by requesting the brief output stage. For example, querying an arXiv ID directly fetches a concise summary, matching your reading depth to the immediate requirement.