deep-research

Conducts structured five-layer research on markets, competitors, regulations, and products with synthesis and red-team validation.

97|4|Updated Mar 31, 2026
One-click install
npx skills add https://github.com/DishantPal/deep-research-skill --skill deep-research-dishantpal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/DishantPal/deep-research-skill/tree/main/deep-research
Command: npx skills add https://github.com/DishantPal/deep-research-skill --skill deep-research-dishantpal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Most AI research outputs are shallow, covering only easily searchable surface information and generic overviews. This Skill enforces a structured five-layer research methodology with analytical frameworks, synthesis techniques, and quality gates so research deliverables contain non-obvious insights, sourced claims, and decision-ready analysis. ## Core Features & Use Cases - Five-Layer Research Methodology: Executes Surface Scan, Structure & Landscape, Depth Dive, Adjacent Opportunities, and Horizon Pointers layers, each with explicit quality gates. - 12 Analytical Frameworks: Applies PESTEL, Porter's Five Forces, Jobs-to-be-Done, Competitive Positioning Map, SWOT, and more with full application instructions and selection logic. - Synthesis & Red-Team Engine: Triangulates sources, detects convergence/divergence/silence patterns, calibrates confidence levels, and stress-tests conclusions before delivery. - 6 Topic Playbooks: Provides step-by-step execution sequences for Market Entry, Product Research, Competitive Analysis, Regulatory, Problem-Solving, and Service Expansion research. - Use Case: Ask "What does the competitive landscape look like for telehealth GLP-1 services?" and receive a structured research brief with a TL;DR, landscape map, practitioner-level depth findings, counterarguments, decision points, and confidence-rated sources. ## Quick Start Ask Claude to research a market, competitor set, regulation, or product decision, for example: "Research the utility discount market for disability placard holders and tell me whether it's worth entering."

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I do deep market research with Claude?▼

Trigger the skill with research-intent language like "research the market for X" or "dig into how Y works." It executes a five-layer methodology covering surface facts, landscape structure, practitioner-level depth, adjacent opportunities, and horizon pointers, then delivers a synthesized research brief.

What analytical frameworks does this research skill use?▼

It includes 12 frameworks with full application instructions: PESTEL, Porter's Five Forces, Jobs-to-be-Done, Competitive Positioning Map, Business Model Canvas, SWOT, Value Proposition Canvas, Technology Adoption Lifecycle, Regulatory Landscape Mapping, Market Segmentation, Root Cause Analysis, and Analogous Industry Scan.

When does the skill ask questions before researching?▼

Guided mode triggers when the prompt is vague, broad, or could go in multiple directions, asking 3-5 sharp scoping questions first. Autonomous mode executes immediately when the prompt is specific and detailed. Ambiguous cases default to at most 2 quick questions.

How does the skill avoid shallow research output?▼

It enforces quality gates per layer, requires at least 3 non-obvious insights in the depth section, mandates a red-team step that challenges emerging conclusions, and tags claims with confidence levels based on source triangulation.

What types of research topics are supported?▼

Six playbooks cover market entry, product and feature research, competitive analysis, regulatory and compliance research, problem-solving, and service expansion. Topics spanning multiple categories combine the relevant playbooks.

What are the limitations of AI-driven deep research?▼

Output quality depends on available public sources; gaps in data are flagged in the Sources & Confidence section rather than hidden. Private company financials and unenforced regulatory realities may require manual follow-up such as practitioner interviews.