semantic-scholar

Search academic papers via the Semantic Scholar API and extract full text with GROBID.

10|Updated Jul 22, 2025
One-click install
npx skills add https://github.com/bdambrosio/Cognitive_workbench --skill semantic-scholar
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: semantic-scholar
Source: https://github.com/bdambrosio/Cognitive_workbench/tree/main/src/tools/semantic-scholar
Command: npx skills add https://github.com/bdambrosio/Cognitive_workbench --skill semantic-scholar

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill helps researchers and students quickly find and access academic papers, overcoming the challenge of sifting through vast amounts of literature.

Core Features & Use Cases

  • Academic Paper Search: Leverages the Semantic Scholar API to find relevant research papers based on a query.
  • Full Text Extraction: When available, it extracts the full text of papers using GROBID, providing more than just abstracts.
  • Use Case: A researcher needs to find the latest papers on "reinforcement learning in robotics." This Skill can search for them, and if PDFs are available, extract the full content for deeper analysis.

Quick Start

Use the semantic-scholar skill to search for papers on "transformer architecture" and synthesize the key contributions.

Frequently Asked Questions about semantic-scholar

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

FAQPage Schema
How do I search academic papers and extract the full text for a literature review?

Searching academic papers requires the Semantic Scholar API to find relevant research, returning abstracts by default or full text content via GROBID parsing when PDFs are available, outputting structured metadata for further analysis.

Do I need an API key to find research papers using Semantic Scholar?

Yes, finding research papers requires a SEMANTIC_SCHOLAR_API_KEY. Optionally, a grobid_url is needed if you want to extract the full text of academic papers rather than just abstracts.

What is the best way to get structured metadata and text from academic papers?

Getting structured metadata and text from academic papers involves querying the Semantic Scholar API and using GROBID for full text extraction, yielding structured outputs of paper metadata and text for synthesis.

Can I extract full text from academic papers or only abstracts?

You can extract full text from academic papers if a GROBID URL is configured. Without GROBID, the Semantic Scholar API search returns paper abstracts instead of the complete text content.

What are the limitations of using GROBID for academic paper search?

A limitation of using GROBID for academic paper search is that full text extraction only works when PDFs are available. If full text is inaccessible, the search falls back to returning abstracts via the Semantic Scholar API.