deep-research

Enforce search-first web research with tiered verification and primary source citation.

6|Updated Apr 15, 2026
One-click install
npx skills add https://github.com/luispabon/ai-bits-and-pieces --skill deep-research-luispabon
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/luispabon/ai-bits-and-pieces/tree/main/skills/deep-research
Command: npx skills add https://github.com/luispabon/ai-bits-and-pieces --skill deep-research-luispabon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the problem of unreliable or hallucinated AI answers by enforcing a rigorous, search-first methodology that requires verifiable evidence for every claim.

Core Features & Use Cases

  • Search-First Workflow: Mandates web searches before generating any content, ensuring answers are grounded in real-time data.
  • Configurable Depth: Offers three levels of research intensity (S1-S3) to balance speed with comprehensive analysis.
  • Use Case: Use this skill to perform a competitive analysis on a new market entrant, requiring a summary of findings, identification of conflicting data, and a confidence score based on primary sources.

Quick Start

Use the deep-research skill with strength 3 to investigate the current state of quantum computing adoption in the financial sector.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I perform evidence-based web research without getting AI hallucinations?

Evidence-based web research prevents hallucinations by enforcing a search-first methodology that mandates web searches and requires verifiable primary sources for every generated claim. This structured approach ensures answers are grounded in real-time data rather than fabricated content.

What is the best way to conduct competitive analysis with conflicting data sources?

Conducting competitive analysis with conflicting sources requires a multi-step verification process that identifies conflicting data, resolves discrepancies, and calculates a confidence score based on primary source citations. This ensures high-stakes inquiries yield reliable summaries.

How do I configure research depth for complex information gathering tasks?

Configure research depth using a tiered strength system offering three levels of intensity (S1-S3) to balance speed with comprehensive analysis. Higher strength levels manage increased query volume and deeper analysis for complex, high-stakes information gathering tasks.

Does structured web research work for tracking fast-moving technology adoption trends?

Structured web research works for tracking technology adoption by mandating real-time web searches before generating any content. This search-first workflow ensures investigations into fast-moving sectors like quantum computing reflect the most current verifiable evidence available.

When should I not use a search-first research approach?

Avoid using a search-first research approach for tasks requiring instant synthesis without web access, as the methodology enforces mandatory multi-step verification and query volume. It is designed for high-stakes inquiries requiring primary source citation rather than rapid, unverified responses.