skills.rest
cardinalconseils
cardinalconseilsCOMMUNITY
Explore all skills by this author →
GitHub

deep-research

Automate recursive multi-source research and generate cited intelligence reports.

3|Updated Mar 16, 2026
Search Tags:#research#citations#deep-research#intelligence#multi-hop#contradictions#source-management
One-click install
npx skills add https://github.com/cardinalconseils/claude-starter --skill deep-research-cardinalconseils
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/cardinalconseils/claude-starter/tree/main/skills/deep-research
Command: npx skills add https://github.com/cardinalconseils/claude-starter --skill deep-research-cardinalconseils

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enables iterative, multi-source investigation to produce structured intelligence reports with confidence scores and source attributions.

Core Features & Use Cases

  • Recursive topic exploration, with configurable sources and depth, to surface actionable insights.
  • Sub-topic discovery, contradiction detection, and synthesis into a final report with traceable citations.
  • Output artifacts under .research/ (report.md, sources.md, matrix.md, and raw results) for easy review and archival.
  • Interactive and automated modes to run analyses, iterations, and follow-up queries as needed.

Quick Start

Run a topic through the deep research workflow to generate a comprehensive intelligence report with source citations.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How does multi-hop research generate intelligence reports with traceable citations?▼

Recursive research automates multi-source exploration by iteratively discovering sub-topics, detecting contradictions, and synthesizing traceable citations into a final structured intelligence report.

How do I automate market analysis with contradiction tracking and confidence scoring?▼

You configure depth levels and sources to run automated market analysis, which recursively explores data, tracks contradictions, assigns confidence scores, and outputs a synthesized report.

Can I run interactive follow-up queries to iterate on strategic intelligence findings?▼

Yes, interactive and automated modes support running analyses, iterations, and follow-up queries to refine strategic intelligence findings based on your ongoing investigation needs.

What output artifacts are generated when doing recursive multi-source research?▼

Recursive research generates output artifacts under .research/ including report.md, sources.md, matrix.md, and raw results to facilitate easy review and archival of your intelligence findings.

Does deep research work for technology evaluations requiring sub-topic discovery and guardrails against failures?▼

Yes, deep research applies to technology evaluations by enabling recursive sub-topic discovery and applying configurable guardrails against failures to ensure reliable strategic intelligence synthesis.

What are the limitations of automated research for generating structured intelligence reports?▼

Automated research depends on configurable sources and depth levels, meaning output quality is bounded by source accessibility and the applied guardrails against failures during recursive exploration.

Related Skills

Explore More Education & Research Skills→
cklxxcklxx
11

deep-research

Convert research questions into structured, sourced conclusions with evidence tracking.

Community
Basic
kmshihab7878kmshihab7878
6

market-competitors

Identify competitors and analyze their positioning for a target website.

Community
Advanced
henrique-simoeshenrique-simoes
13

desk-research

Summarize research literature into source-backed Nuggets, Facts, Insights, and Recommendations.

Community
Intermediate
jgwilljgwill
1

deep-research

Decompose topics into MECE angles and orchestrate parallel Opus sub-agents for research synthesis.

Community
Advanced