research

Routes research tasks to optimal backends including subagents and CLI agents.

Updated Aug 23, 2026

Search Tags:#code analysis #research #web search #subagent #company intel #cli agents

Ready to run

npx skills add https://github.com/EtanHey/golems --skill research-etanhey

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: research
Source: https://github.com/EtanHey/golems/tree/main/skills/golem-powers/research
Command: npx skills add https://github.com/EtanHey/golems --skill research-etanhey

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill streamlines complex research tasks by intelligently routing them to the most efficient backend, whether it's an internal subagent, a command-line AI, or a specialized researcher.

Core Features & Use Cases

Multi-Backend Routing: Selects the best tool (subagent, CLI, etc.) based on research type and cost.
Versatile Research: Handles web research, company intel, code patterns, academic papers, and code audits.
Use Case: When asked to investigate a new technology, the skill will first check existing knowledge, then dispatch the task to a specialized researcher subagent, saving time and resources.

Quick Start

Use the research skill to investigate the topic of quantum computing.

Frequently Asked Questions about research

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

FAQPage Schema

How do I orchestrate deep web research using multiple AI agents?▼

Deep web research orchestration routes tasks to optimal backends like internal subagents, CLI agents such as Gemini or Cursor, or a dedicated researcher subagent. It selects the most efficient tool based on research type and cost, streamlining complex investigations.

Can I use CLI agents like Gemini and Cursor for company intelligence gathering?▼

Yes, you can gather company intelligence by routing profiling tasks to specialized CLI agents. The system supports dedicated company profiling modes, dispatching research to command-line AI tools to retrieve and structure relevant corporate data.

What is the best way to research code patterns before starting an implementation?▼

Pre-implementation research is handled by checking existing knowledge bases first, then dispatching code pattern analysis to a specialized researcher subagent. This approach identifies optimal code structures and audits existing patterns before development begins.

Does this research orchestrator support quick lookups as well as deep dives?▼

Yes, the research orchestrator supports both quick lookups and deep dives. It intelligently routes tasks based on the required depth, selecting faster internal subagents for quick information retrieval and dedicated researcher subagents for comprehensive deep dives.

How does knowledge retrieval and storage work during code analysis?▼

Knowledge retrieval and storage during code analysis integrates with BrainLayer. The orchestrator checks this existing knowledge base before dispatching tasks, saving time and resources by leveraging previously stored research and code audit results.