deep-research

Produce multi-source web research reports with inline citations in Markdown.

3|3|Updated Apr 5, 2026
One-click install
npx skills add https://github.com/diegocamara89/ai-skills-hub --skill deep-research-diegocamara89
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/diegocamara89/ai-skills-hub/tree/main/all-skills/deep-research
Command: npx skills add https://github.com/diegocamara89/ai-skills-hub --skill deep-research-diegocamara89

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates time-consuming manual web investigation by orchestrating multi-source searches, deep-reading key documents, and producing a single, evidence-backed report with precise source attribution.

Core Features & Use Cases

  • Multi-source search orchestration: combines web search and semantic tools to gather 15-30 relevant sources across news, papers, forums, and benchmarks.
  • Deep-read synthesis with citations: reads full pages, extracts facts, cross-references claims, and produces structured Markdown reports with inline citations and a sources list.
  • Parallelized subagents for scale: splits broad topics into focused agents (benchmarks, reviews, news) and merges findings into a coherent executive summary.
  • Use Case: perform due diligence on a technology vendor, produce a competitive landscape report, or compile benchmark comparisons with verifiable citations.

Quick Start

Use the deep-research skill to produce a cited research report on "State of AI coding assistants in 2025" covering key tools, benchmarks, pricing, and developer sentiment.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I generate multi-source web research reports with inline citations?

Multi-source web research reports with inline citations are generated by orchestrating parallelized subagents to search the web, deep-read full pages, cross-reference claims, and synthesize findings into a structured Markdown document with verifiable source attribution.

What's the best way to automate due diligence research on a technology vendor?

Automated due diligence is best handled by splitting broad topics into focused agents that gather 15-30 sources across news and benchmarks, cross-reference facts, and merge findings into a coherent executive summary with precise citations.

Do I need a web search tool to produce cited competitive benchmarking reports?

Yes, producing cited competitive benchmarking reports requires at least one configured web search or fetch tool, along with the ability to deep-read full pages and extract facts for cross-referencing.

How does deep-read synthesis work for technical literature reviews?

Deep-read synthesis for technical literature reviews works by reading full pages, extracting factual claims, cross-referencing them across multiple sources, and producing a structured Markdown report with inline citations and a sources list.

Can I compile market analysis reports by merging findings from parallelized research agents?

Yes, you can compile market analysis reports by splitting broad topics into focused parallelized subagents that gather benchmarks, reviews, and news, then merging their findings into a coherent executive summary.

What are the limitations of using automated web research for evidence-backed synthesis?

Automated web research for evidence-backed synthesis is limited by the configured search tool's reach and the ability to deep-read full pages, meaning heavily paywalled or dynamically rendered sources may restrict cross-referencing accuracy.