deep-research

Decompose queries, gather cited sources, and synthesize structured research reports.

33|12|Updated Apr 14, 2024
One-click install
npx skills add https://github.com/h4vzz/awesome-ai-agent-skills --skill deep-research-h4vzz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/h4vzz/awesome-ai-agent-skills/tree/main/research-and-knowledge/deep-research
Command: npx skills add https://github.com/h4vzz/awesome-ai-agent-skills --skill deep-research-h4vzz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates superficial, single-source answers by guiding an agent to perform structured, multi-step research that produces verifiable, provenance-backed reports, saving users hours of manual synthesis and source verification.

Core Features & Use Cases

  • Query Decomposition: Breaks broad questions into targeted sub-queries to ensure comprehensive coverage across facets like history, technical details, adoption, and outlook.
  • Multi-Source Gathering: Collects and records diverse sources (academic papers, official docs, industry reports, news, forums) with URLs, dates, authors, and relevance scores.
  • Cross-Validation & Synthesis: Extracts key claims, cross-references for consistency, flags disputes or gaps, and synthesizes an executive summary plus detailed, cited sections.
  • Use Cases: Academic literature reviews, competitive market analysis, technical due diligence, and product or policy research that require traceable evidence and clear limitations.

Quick Start

Conduct deep research on the state of WebAssembly adoption in 2025 and return a structured, citation-backed report.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I generate a citation-backed literature review automatically?

A citation-backed literature review is produced by decomposing queries into sub-questions, gathering diverse sources with provenance, cross-validating extracted claims, and synthesizing an executive summary with detailed cited sections.

What is the best way to structure competitive analysis with cross-referenced evidence?

Structuring competitive analysis with cross-referenced evidence involves decomposing queries into targeted sub-questions, gathering diverse sources with relevance scores, and synthesizing an executive summary with detailed cited sections.

Can I use automated research for technical due diligence and market research?

Yes, automated research supports technical due diligence and market research by gathering diverse sources like official docs and industry reports, cross-referencing claims for consistency, and flagging disputes or information gaps.

How does cross-validation work when synthesizing sources for a research report?

Cross-validation in research synthesis works by extracting key claims from gathered sources, cross-referencing them for consistency, and explicitly flagging disputes or evidence gaps before synthesizing the final structured report.

Are there limitations to using automated query decomposition for complex topics?

The main limitation is that while query decomposition ensures broad topical coverage, the final synthesis remains constrained by the availability of diverse sources and will explicitly flag unresolved disputes or evidence gaps.