Paper Discovery

Discover and curate scholarly papers across arXiv, PubMed, Semantic Scholar, and OpenAlex.

12|7|Updated Mar 10, 2025
One-click install
npx skills add https://github.com/acertainKnight/project-thoth --skill paper-discovery
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Paper Discovery
Source: https://github.com/acertainKnight/project-thoth/tree/main/src/thoth/.skills/paper-discovery
Command: npx skills add https://github.com/acertainKnight/project-thoth --skill paper-discovery

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Researchers often spend excessive time locating and organizing relevant papers across multiple databases. This Skill streamlines discovery and curation, delivering a targeted set of articles ready for review.

Core Features & Use Cases

  • Cross-source discovery: Search across arXiv, PubMed, Semantic Scholar, and OpenAlex to surface relevant papers.
  • Question-driven curation: Create research questions and retrieve focused results for literature reviews.
  • Results management: Review, filter, and export summaries or metadata for downstream analysis.

Quick Start

Step 1: Create a research question create_research_question( title="Your topic in 1-2 sentences", keywords=["topic keyword"], sources=["arxiv", "pubmed", "semantic_scholar", "openalex"], max_papers=25, relevance_threshold=0.7 )

Step 2: Run discovery run_discovery_for_question(question_id="[from step 1]")

Step 3: List results list_articles(limit=20, sort_by="relevance")

Frequently Asked Questions about Paper Discovery

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

FAQPage Schema
How do I search for research papers across multiple databases like PubMed and arXiv?

Paper Discovery searches across arXiv, PubMed, Semantic Scholar, and OpenAlex simultaneously. Define a research question with keywords and sources, set your relevance threshold and paper limit, then run discovery to surface relevant papers from all selected databases at once.

Can I use paper discovery tools to organize a literature review?

Yes. Paper Discovery streamlines literature reviews by letting you create research questions, run cross-source discovery, filter results by relevance, and export summaries or metadata for downstream analysis and organization.

What's the best way to filter and manage research paper results from multiple sources?

After discovery completes, list and review articles sorted by relevance, apply filtering to narrow results, then export structured summaries and metadata. This workflow moves papers from raw search results to curated, analyzable datasets.

Does paper discovery work with OpenAlex and Semantic Scholar?

Yes. Paper Discovery integrates OpenAlex, Semantic Scholar, arXiv, and PubMed. You select which sources to query when creating a research question, and discovery pulls results from all chosen databases.

How do I structure a research question to get focused paper results?

Create a research question by specifying a topic title, keywords relevant to your area, target sources, maximum paper count, and a relevance threshold. This structured approach ensures discovery returns targeted results aligned with your research scope.

What happens if my paper discovery returns too many or too few results?

Adjust your relevance threshold, refine keywords, reduce or expand sources, or change max_papers count. Lower thresholds and broader keywords increase result volume; higher thresholds and precise keywords narrow results to higher-confidence matches.