research

Gather, verify, and synthesize diverse sources into a Markdown research brief.

Updated Mar 18, 2026
One-click install
npx skills add https://github.com/arvindbattula/my-agent-config --skill research-arvindbattula
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/arvindbattula/my-agent-config/tree/main/skills/research
Command: npx skills add https://github.com/arvindbattula/my-agent-config --skill research-arvindbattula

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Systematically gathers, verifies, and synthesizes information from diverse sources to deliver a consolidated research brief.

Core Features & Use Cases

  • Phase-based Discover → Deepen → Challenge with adversarial verification to close open questions and deliver a structured brief.
  • Tiered engine routing (WebSearch, SerpAPI indexes) to surface diverse perspectives and reduce single-source bias.
  • Automated source extraction, synthesis, and question tracking suitable for market analysis, technology evaluation, literature reviews, and competitive analysis.
  • Final brief in Markdown with an explicit Open Questions section for ongoing investigations.

Quick Start

Provide a topic and constraints to start Discover → Deepen → Challenge and generate a final research brief.

Frequently Asked Questions about research

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

FAQPage Schema
How do I gather and synthesize information from diverse online sources into a single research brief?

Multi-source research systematically gathers and verifies information using iterative convergence and adversarial verification to deliver a consolidated Markdown brief with an explicit open questions section.

What is adversarial verification and how does it help close open questions in market analysis?

Adversarial verification challenges synthesized findings during the Deepen phase to identify knowledge gaps, ensuring open questions are explicitly tracked and addressed before finalizing the research brief.

How do I generate a literature review or competitive analysis without single-source bias?

Tiered engine routing uses WebSearch and SerpAPI indexes to surface diverse perspectives, automatically extracting and synthesizing sources to reduce single-source bias in market and technology evaluations.

Can I use WebSearch and SerpAPI-based engines together for technology evaluation research?

Yes, tiered engine routing supports both WebSearch and SerpAPI-based indexes when available, surfacing diverse perspectives and reducing single-source bias for technology evaluation and literature reviews.

What's the best way to track unresolved knowledge gaps when doing multisource research?

The final Markdown brief includes an explicit Open Questions section, automatically tracking unresolved knowledge gaps suitable for ongoing investigations in market analysis and expert opinion synthesis.

Does automated research synthesis handle search engine failures during source gathering?

Yes, the multisource research process includes built-in failure handling for tiered engine routing, ensuring the Discover, Deepen, and Challenge phases complete despite search engine failures.