deep-research

Conducts multi-source web research and synthesizes cited findings from crawled pages.

Updated May 17, 2022
One-click install
npx skills add https://github.com/kanade0404/dotfiles --skill deep-research-kanade0404
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/kanade0404/dotfiles/tree/main/.claude-plugin/skills/deep-research
Command: npx skills add https://github.com/kanade0404/dotfiles --skill deep-research-kanade0404

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill turns broad, ambiguous research questions into structured, evidence-based answers drawn from multiple web sources, reducing time spent manually searching and comparing information.

Core Features & Use Cases

  • Multi-source search and synthesis: Combines results from firecrawl and exa across news, official docs, and reputable web sources.
  • Deep reading and citation: Scrapes and crawls key pages to support claims with inline citations and source attribution.
  • Use cases: Competitive analysis, technology evaluation, due diligence, market sizing, and any question that needs a concise report backed by evidence.

Quick Start

Ask the assistant to research your topic in depth, gather supporting sources, and return a cited summary with key findings.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I conduct competitive analysis with cited web sources?

Competitive analysis with cited web sources is conducted by searching multiple reputable sites, scraping key pages for evidence, and synthesizing findings into a report with inline citations. This process separates verified facts from inference using firecrawl or exa MCP access.

What is the best way to gather evidence for technology evaluation?

The best way to gather evidence for technology evaluation is multi-source web research that scrapes official documentation and news. It synthesizes data into an attributed report, ensuring claims are backed by inline citations and clearly separating verified facts from inference.

Does this multi-source research require firecrawl or exa MCP access?

Yes, this multi-source research requires firecrawl or exa MCP access. These dependencies are essential for the assistant to search, scrape, crawl, and cite web sources while producing evidence-based findings for complex questions.

How do I perform due diligence research using web search and synthesis?

Due diligence research using web search and synthesis is performed by querying broad questions across multiple sources, crawling key pages for supporting evidence, and generating a concise report. It attributes claims with inline citations and distinguishes verified facts from inference.

Can I use this for market sizing and evidence-based reporting?

Yes, you can use this for market sizing and evidence-based reporting. It applies multi-source web research to complex questions, scraping and crawling reputable sources to produce concise reports backed by attributed findings and inline citations.

What are the limitations of automated web research and synthesis?

Limitations of automated web research include reliance on firecrawl or exa MCP access for searching and scraping. It requires clear separation of verified facts from inference, meaning ambiguous queries may yield less concrete evidence for complex research tasks.