research

Conduct web research and validate findings through multi-perspective AI analysis.

13|3|Updated Mar 2, 2026
One-click install
npx skills add https://github.com/phrazzld/agent-skills --skill research-phrazzld
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/phrazzld/agent-skills/tree/main/core/research
Command: npx skills add https://github.com/phrazzld/agent-skills --skill research-phrazzld

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill streamlines web research by intelligently delegating tasks to various AI models and validating information from multiple perspectives, ensuring comprehensive and reliable insights.

Core Features & Use Cases

  • Web Search & Deep Dive: Perform targeted web searches, news aggregation, and in-depth document retrieval.
  • Multi-AI Delegation: Orchestrate tasks across different AI agents (like Codex or Gemini) for specialized work.
  • Thinktank Validation: Gather diverse expert opinions to validate findings and reach consensus.
  • Session Analysis: Introspect past interactions to identify usage patterns and areas for improvement.
  • Use Case: When researching a complex technical topic, use /research to gather initial information, then /research delegate to have Codex implement a related code snippet, and finally /research thinktank to validate the approach with multiple AI perspectives.

Quick Start

Use the research skill to search the web for the latest advancements in quantum computing.

Frequently Asked Questions about research

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

FAQPage Schema
How do I validate technical research using multiple AI agents?

You can validate technical research by delegating tasks to various AI agents and gathering diverse expert opinions to reach consensus. This multi-perspective analysis ensures comprehensive and reliable insights for complex technical investigations.

What is the best way to orchestrate web search and AI delegation for complex investigation?

The best way to orchestrate web search and AI delegation is to use a skill that performs targeted web searches and delegates specialized work across different AI models. This approach streamlines intelligence gathering and deep document retrieval.

Can I use local script execution to coordinate AI models for competitive analysis?

Yes, you can use local script execution to coordinate AI models for competitive analysis. This orchestration requires integration with various AI providers to execute specialized tasks and validate gathered information effectively.

Does multi-perspective analysis work for synthesizing knowledge from web search?

Multi-perspective analysis works effectively for synthesizing knowledge from web search by gathering diverse expert opinions to validate findings. This consensus mechanism ensures the synthesized information is reliable and comprehensive.

Are there limitations when delegating tasks to AI agents for technical investigation?

Limitations when delegating tasks to AI agents for technical investigation include the dependency on integrating with various external AI providers and requiring local script execution for the orchestration to function properly.

How do I analyze past sessions to improve web research workflows?

You can analyze past sessions to improve web research workflows through session analysis, which introspects past interactions to identify usage patterns. This process highlights areas for improvement in your information retrieval strategies.