deep-research

Execute structured multi-phase research with recursive evidence tracking and source verification.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/arkadeepduttatrivago/trv-robin-aios --skill deep-research-arkadeepduttatrivago
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/arkadeepduttatrivago/trv-robin-aios/tree/main/AIOS/skills/deep-research
Command: npx skills add https://github.com/arkadeepduttatrivago/trv-robin-aios --skill deep-research-arkadeepduttatrivago

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires WebSearch, WebFetch, internal-context, source-hierarchy, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill addresses the challenge of conducting thorough, multifaceted research by providing a systematic framework for collecting, analyzing, and synthesizing multi-source information to produce actionable, confidence-rated findings.

Core Features & Use Cases

  • Structured Methodology: Guides users through four rigorous research phases—scope, execution, analysis, and synthesis—ensuring depth and coverage.
  • Recursive & Cyclical Exploration: Facilitates iterative searches and evidence chain tracing, improving source credibility and confidence levels.
  • Multi-Source Source Tracking: Maintains a comprehensive source registry with URLs and confidence ratings for transparency and verification.
  • Versatile Output Formats: Supports executive summaries, key findings, full reports, comparison matrices, and action plans tailored to decision-making needs.
  • Use Case: Ideal for competitive intelligence, industry benchmarking, macro trend analysis, or corporate strategic research requiring high reliability.

Quick Start

Follow the step-by-step guide in the documentation to plan your research, define questions, decompose queries, and utilize recursive exploration for comprehensive results.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I conduct multi-source analysis for strategic decision making?

Multi-source analysis is conducted through a structured methodology covering scope, execution, analysis, and synthesis phases. This process integrates external searches with internal context, tracing evidence chains to generate confidence-rated insights for strategic decisions.

What is recursive evidence tracking in deep research?

Recursive evidence tracking is an iterative search mechanism that traces evidence chains across multiple sources. It improves source credibility by maintaining a comprehensive registry with URLs and confidence ratings for continuous verification during research.

How do I verify sources during competitive intelligence research?

Source verification is achieved by maintaining a comprehensive source registry with URLs and confidence ratings. The Skill integrates external searches with internal organizational context to ensure high-quality outputs suitable for competitive intelligence.

Can I use internal organizational context with external searches for industry benchmarking?

Internal organizational context is integrated directly with external web searches during the research workflow. This combination ensures comprehensive coverage and high reliability for industry benchmarking and macro trend analysis.

What's the best way to structure a comprehensive research report?

The best way to structure a research report is following the four-phase methodology: scope, execution, analysis, and synthesis. It supports versatile formats including executive summaries, key findings, comparison matrices, and action plans.

When should I not use a structured multi-phase research workflow?

A structured multi-phase research workflow is not suitable for simple, single-query information retrieval tasks. It is designed for comprehensive organizational and industry topics requiring recursive exploration and high reliability, not quick lookups.