deep-research

Run iterative research loops with falsification and maintain audit artifacts.

17|2|Updated Dec 15, 2025
One-click install
npx skills add https://github.com/VonEquinox/WebSearchMCP --skill deep-research-vonequinox
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/VonEquinox/WebSearchMCP/tree/main/deep-research
Command: npx skills add https://github.com/VonEquinox/WebSearchMCP --skill deep-research-vonequinox

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill turns vague or contested questions into an auditable, source-backed investigation by running repeated broaden/deepen/reframe/falsify loops and maintaining explicit artifacts so conclusions are traceable and stress-tested.

Core Features & Use Cases

  • Structured iterative workflow: Orchestrates repeated cycles of broadening (discovery), deepening (primary-source verification), reframing, and falsification until insights stabilize.
  • Living artifacts for auditability: Initializes and maintains a research-plan, evidence-ledger, hypothesis-map, contradiction-matrix, iteration-log, and decision-log to capture provenance and decisions.
  • Explicit falsification and boundary analysis: Actively seeks counterevidence, documents conditions where claims hold or fail, and records reconciliation status and stop criteria.
  • Use cases: exploratory academic literature reviews, technical failure-mode investigations, policy impact analysis, and any open-ended research requiring rigorous hypothesis testing and traceable evidence.

Quick Start

Investigate the question "[insert topic]" with iterative broaden→deepen→reframe→falsify loops and produce a sourced research report plus the research-plan, evidence-ledger, hypothesis-map, contradiction-matrix, iteration-log, and decision-log.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I conduct an iterative literature review with competing hypotheses?

An iterative literature review applies repeated broaden, deepen, reframe, and falsify loops to investigate competing hypotheses. It maintains a hypothesis-map and evidence-ledger to trace primary sources, ensuring conclusions are auditable and stress-tested against counterevidence.

What is falsification in open-ended academic research?

Falsification in open-ended research actively seeks counterevidence to test boundary conditions where claims hold or fail. It documents reconciliation status within a contradiction-matrix, ensuring that conclusions are evidence-led rather than simply confirmed by supportive sources.

How do I verify primary sources for a policy impact analysis?

Verifying primary sources for policy impact analysis requires an evidence-ledger that captures provenance and decisions. By deepening inquiries through iterative loops, you can prioritize verified primary sources and record reconciliation status within a structured research-plan.

Can I use iterative research for technical failure-mode investigations?

Iterative research suits technical failure-mode investigations by applying explicit falsification and boundary analysis across ambiguous exploratory scenarios. It maintains living artifacts like a decision-log and iteration-log to capture provenance and ensure traceable evidence.

What is the best way to structure an auditable research narrative?

The best way to structure an auditable research narrative is to maintain living artifacts including a research-plan, evidence-ledger, hypothesis-map, and contradiction-matrix. These components capture provenance and decisions, turning vague questions into sourced, traceable conclusions.

When should I not use an iterative open-ended research approach?

An iterative open-ended research approach should not be used for straightforward factual lookups or single-answer questions. It is designed for ambiguous, contested topics requiring rigorous hypothesis testing, boundary-condition analysis, and repeated falsification loops to stabilize insights.