deep-research

Enhance research prompts and execute web-backed deep research with timestamped reports.

3|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/pinion05/skills --skill deep-research-pinion05
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/pinion05/skills/tree/main/deep-research
Command: npx skills add https://github.com/pinion05/skills --skill deep-research-pinion05

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This skill enables comprehensive, internet-enabled research by orchestrating prompt enhancement, saving research parameters for reproducibility, and executing deep research with web search capabilities. It helps users who need structured, in-depth analysis without manual prompt tuning.

Core Features & Use Cases

  • Smart prompt enhancement: detects vague prompts and asks clarifying questions (2-3 targeted questions) to sharpen scope.
  • Reproducible research: saves enhanced prompts to timestamped files and records parameters for auditability.
  • End-to-end execution: runs a deep research pipeline that leverages web search and returns a final Markdown report with sources.
  • Use cases: lengthy market analyses, technology trend studies, threat-models, or any topic requiring structured, web-backed research.

Quick Start

Provide a research prompt to the skill (e.g., "What are the latest AI governance best practices?"). The system will assess prompt quality, optionally ask clarifying questions, save the enhanced prompt to a timestamped file, run the deep research process, and present the final report with sources. If you want to skip enhancement, pass --no-enhance to the orchestrator.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I automate in-depth research with web search and reproducible parameters?

Automate in-depth research by providing a prompt to an orchestrator that enhances scope, saves timestamped parameters, and executes a web-backed pipeline to generate a final Markdown report with sources.

Can I skip prompt enhancement when running a deep research pipeline?

Yes, you can skip prompt enhancement by passing the --no-enhance flag to the orchestrator, which bypasses clarifying questions and executes the deep research pipeline directly.

Does this deep research tool work for market studies and technology trend analysis?

Yes, the deep research tool works for market studies, technology trend analysis, and threat-models by running structured, web-backed analyses that return comprehensive Markdown reports with sources.

How does smart prompt enhancement improve vague research queries?

Smart prompt enhancement detects vague research queries and asks 2-3 targeted clarifying questions to sharpen scope, ensuring the deep research pipeline produces highly relevant and structured web-backed analyses.

Do I need an OpenAI model to execute web search and generate research reports?

Yes, you need a configurable OpenAI model to execute the end-to-end deep research pipeline, which leverages web search capabilities to return a final Markdown report with sources.

What is the best way to ensure reproducibility when conducting automated web research?

Ensure reproducibility in automated web research by using a system that saves enhanced prompts to timestamped files and records all research parameters, creating an auditable trail for future reference.