deep-research

Conduct multi-source web research and synthesize cited findings into structured reports.

Updated Apr 14, 2026
One-click install
npx skills add https://github.com/NS-Vandana/Med-Save-Hub --skill deep-research-ns-vandana
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/NS-Vandana/Med-Save-Hub/tree/main/secondary_skills/deep-research
Command: npx skills add https://github.com/NS-Vandana/Med-Save-Hub --skill deep-research-ns-vandana

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? Many complex questions require more than a few web searches: fragmented sources, unverified claims, and lack of synthesis make it hard to produce trustworthy, actionable conclusions. This Skill provides a repeatable workflow to gather, evaluate, cross-reference, and synthesize information from many sources into a single, cited report.

## Core Features & Use Cases

  • Topic decomposition: Breaks broad questions into 5 focused research areas to ensure comprehensive coverage.
  • Parallel source discovery: Launches concurrent subagents to run diverse web searches and fetch full articles for deeper inspection.
  • Source evaluation & synthesis: Assesses authority, currency, bias, and accuracy, cross-references claims, and organizes findings into a structured report with citations.
  • Use cases: Literature reviews, market or technology analyses, policy impact studies, verification of contested claims, and in-depth educational research.

### Quick Start Ask the skill to perform a deep dive into your topic, decompose it into focused areas, run parallel searches, evaluate credibility, and return a structured report with explicit source citations and limitations noted.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I conduct a literature review with citations from multiple web sources?

To conduct a literature review with citations, decompose your broad topic into focused subareas, run parallel web searches, evaluate source credibility, cross-reference claims, and synthesize findings into a structured report with explicit source URLs.

What is the best way to verify contested claims found during web research?

The best way to verify contested claims is to fetch full articles from multiple sources, assess their authority and bias, cross-reference the claims against each other, and document any limitations in a structured synthesis report.

How do I synthesize fragmented market analysis data into a single cited report?

You synthesize fragmented market analysis data by launching concurrent web searches across diverse sources, evaluating the accuracy and currency of each, cross-referencing the data, and formatting the results into a cited report.

Can I perform a comprehensive technology analysis without manually checking each source?

Yes, you can perform a comprehensive technology analysis by decomposing the topic into five focused areas, running parallel subagents to fetch articles, and automatically evaluating source authority and bias.

What are the limitations of automated source evaluation for deep dive research?

Limitations of automated source evaluation include potential gaps in comprehensive coverage and the necessity to document limitations explicitly, as cross-referencing claims relies on available web sources and their stated accuracy.

When do I need to decompose a research question into focused subareas?

You need to decompose a research question into focused subareas when facing complex topics requiring deep dives, ensuring comprehensive coverage by running parallel web searches and cross-referencing multiple sources.