research

Coordinate parallel multi-source web research across multiple AI agents.

9|1|Updated Apr 16, 2026
One-click install
npx skills add https://github.com/CarbeneAI/Forge --skill research-carbeneai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/CarbeneAI/Forge/tree/main/.claude/skills/Research
Command: npx skills add https://github.com/CarbeneAI/Forge --skill research-carbeneai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Centralizes and accelerates complex web and content research by coordinating multiple researcher agents, fetching difficult-to-access content, and synthesizing findings into actionable reports so users avoid manual, slow, and fragmented investigations.

Core Features & Use Cases

  • Parallel Multi-Agent Research: Launches Claude, Perplexity, Gemini, Grok, Ollama and other researcher agents in parallel with configurable modes (quick, standard, extensive) to maximize coverage and speed.
  • Intelligent Retrieval & Escalation: Three-layer content retrieval (WebFetch → BrightData → Apify) for handling blocked/CAPTCHA-protected sites and complex JavaScript pages.
  • Fabric Pattern Selection & Extraction: Auto-selects from 242+ Fabric patterns for summarization, wisdom extraction, threat modeling, and specialized content transformations.
  • Use Case: Rapidly produce a sourced synthesis of current events, extract insights from YouTube/papers, and generate interview briefs or threat models across multiple sources.

Quick Start

Run a parallel multi-source research on "quantum computing breakthroughs" and synthesize findings across available researcher agents.

Frequently Asked Questions about research

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

FAQPage Schema
How do I conduct parallel multi-source web research and synthesize findings from multiple AI agents?

Parallel multi-source web research coordinates multiple researcher agents like Claude, Perplexity, and Gemini to fetch, extract, and synthesize information concurrently. It decomposes queries and orchestrates agents with timeouts to maximize coverage and speed.

What is the best way to scrape blocked or CAPTCHA-protected web pages for content extraction?

Scraping blocked or CAPTCHA-protected pages uses a three-layer retrieval fallback system: WebFetch attempts initial access, then escalates to BrightData, and finally Apify to handle complex JavaScript and protected sites.

Can I use Fabric patterns for summarization and knowledge extraction from YouTube videos or papers?

Yes, Fabric pattern selection automatically chooses from over 242 patterns for knowledge extraction, summarization, and specialized content transformations. This applies to extracting insights from videos, papers, and multi-source articles.

Does multi-agent research work with local models like Ollama alongside cloud APIs?

Multi-agent research supports local models like Ollama alongside cloud agents including Grok, Gemini, and Claude. It orchestrates these agents in parallel across quick, standard, or extensive modes depending on required coverage.

How do I generate a sourced synthesis of current events across multiple researcher agents?

Generating a sourced synthesis involves launching parallel agents to investigate the topic, applying three-layer retrieval for content access, and using Fabric patterns to synthesize findings into actionable reports.

What are the limitations of using orchestrated multi-agent research for complex web scraping?

Limitations of orchestrated multi-agent research include managing agent timeouts during parallel execution and handling retrieval failures. The three-layer fallback mitigates access issues, but complex JavaScript pages may still require specific API integrations.