expert-research

Runs multi-perspective web research with fact-checking, critic review, and decision matrices.

4|Updated Apr 17, 2026
One-click install
npx skills add https://github.com/hint-shu/deep-research --skill expert-research-hint-shu
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: expert-research
Source: https://github.com/hint-shu/deep-research/tree/main/skills/expert-research
Command: npx skills add https://github.com/hint-shu/deep-research --skill expert-research-hint-shu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? High-stakes decisions like technology migrations or architecture choices need more than a quick web search — they require verified claims, multiple viewpoints, and adversarial review, which ad-hoc research rarely delivers. ## Core Features & Use Cases - Multi-Perspective Search: Researches the topic from proponent, critic, and neutral angles across 40-60 sources to counter selection bias. - Fact-Checking & Critic Agent: Verifies the top 5 critical claims against independent sources and runs an adversarial sub-agent review of the report. - Decision Matrix & Executive Summary: Produces a weighted decision matrix for choice queries plus a 500-word stakeholder summary and PDF export. - Use Case: When deciding whether to migrate from one framework to another, invoke this skill to get a 2000-3000 word report with verified claims, counterpoints, and a scored comparison of options. ## Quick Start Run /expert-research with your strategic question, for example asking whether your team should migrate from one technology stack to another.

Frequently Asked Questions about expert-research

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

FAQPage Schema
How do I research a technology migration decision with AI?▼

Invoke /expert-research with your migration question. It runs a deep-research baseline, then adds proponent, critic, and neutral search angles, fact-checks critical claims, and delivers a report with a weighted decision matrix comparing your options.

What is the difference between deep-research and expert-research?▼

Expert-research (L3) composes on top of deep-research (L2) and adds multi-perspective search, a fact-check pass on the top 5 claims, an adversarial critic agent, a second reflection loop, and a decision matrix. It uses 40-60 sources versus L2's 20-30.

How long does expert-level research take and how many sources does it use?▼

A typical run takes about 20-25 minutes and gathers 40-60 sources, consuming roughly 150 Tavily search credits. The output is a 2000-3000 word report plus a 500-word executive summary and PDF export.

Does the research report verify its claims against independent sources?▼

Yes. The fact-check step identifies the top 5 critical claims and verifies each against at least two independent sources, labeling them CONFIRMED, DISPUTED, or UNVERIFIED. Disputed claims are downgraded or caveated in the final report.

When should I not use expert-level research?▼

Skip it for simple fact-checks or quick lookups — use lighter tiers like quick-research or research instead. Expert-research is designed for strategic decisions where being wrong costs significant time or money, not for routine questions.