research

Aggregate evidence from arXiv, academic literature, web sources, and practitioner material into a RESEARCH.md artifact.

1|Updated Jul 1, 2026
One-click install
npx skills add https://github.com/marcuskrogh/skills --skill research-marcuskrogh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/marcuskrogh/skills/tree/main/skills/research
Command: npx skills add https://github.com/marcuskrogh/skills --skill research-marcuskrogh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill solves the problem of fragmented or shallow research by providing a structured, multi-axis framework to gather evidence across preprints, formal literature, web sources, and practitioner material.

Core Features & Use Cases

  • Multi-Axis Discovery: Systematically covers arXiv preprints, peer-reviewed journals, web documentation, and informal engineering discourse.
  • Evidence Synthesis: Generates a comprehensive RESEARCH.md artifact that maps findings to specific sources without prematurely locking in product decisions.
  • Use Case: Use this skill when you need to perform a state-of-the-art survey on a technical topic, such as comparing different diffusion model architectures for protein design, to inform subsequent definition or modeling phases.

Quick Start

Invoke the research skill to perform a comprehensive multi-axis investigation on the topic of transformer efficiency in large language models.

Frequently Asked Questions about research

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

FAQPage Schema
How do I gather evidence from arXiv preprints and formal literature for a technical feasibility study?

To gather evidence for a technical feasibility study, this skill performs a multi-axis investigation by aggregating information from arXiv preprints, formal academic literature, web discovery, and practitioner-focused sources. It systematically covers peer-reviewed journals and informal engineering discourse.

What is the best way to perform a state-of-the-art survey on a technical topic like transformer efficiency?

The best way to perform a state-of-the-art survey is to use a structured multi-axis framework that maps findings to specific sources. This skill synthesizes evidence across preprints, web documentation, and practitioner material to generate a comprehensive RESEARCH.md artifact.

Do I need Python to use arXiv API interaction for literature reviews?

Yes, you need Python for arXiv API interaction when performing broad literature reviews. The skill requires Python to query preprint servers and web search capabilities for broader discovery across academic and practitioner-focused sources.

How do I conduct multi-axis discovery for product definition or modeling tasks?

To conduct multi-axis discovery for product definition, invoke this skill to investigate your topic across formal academic literature and informal engineering discourse. It generates a RESEARCH.md artifact that maps findings to sources without prematurely locking in product decisions.

Why does evidence gathering for modeling tasks require mapping findings to specific sources?

Evidence gathering requires mapping findings to specific sources to avoid prematurely locking in product decisions. By aggregating across formal literature and practitioner material, the skill ensures your modeling tasks are grounded in traceable, comprehensive evidence synthesis.