skills.rest
Kaleb-Rupe
Kaleb-RupeCOMMUNITY
Explore all skills by this author →
𝕏@MightieMagsGitHub

Research

Routes research requests to multi-agent web research workflows with mandatory URL verification.

Updated Jul 10, 2026
Search Tags:#research#web search#multi-agent#content extraction#deep investigation#fabric#url verification
One-click install
npx skills add https://github.com/Kaleb-Rupe/aurora --skill research-kaleb-rupe
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Research
Source: https://github.com/Kaleb-Rupe/aurora/tree/main/claude/skills/Research
Command: npx skills add https://github.com/Kaleb-Rupe/aurora --skill research-kaleb-rupe

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Research agents frequently hallucinate URLs and single-query searches miss depth, so this Skill orchestrates multi-agent web research with enforced URL verification and tiered depth modes. ## Core Features & Use Cases - Tiered Research Modes: Quick (1 agent), Standard (2 agents), Extensive (9 parallel agents across Claude, Gemini, and Grok), and Deep Investigation (iterative, vault-backed) modes matched to request scope. - Content Extraction & Analysis: Extract alpha insights from videos and articles, run 235 Fabric patterns, scrape protected sites via BrightData and Apify fallbacks, and analyze AI industry trends. - Mandatory URL Verification: Every URL returned by research agents is fetched and content-confirmed before delivery, replacing broken links automatically. - Use Case: Ask to investigate a market landscape and the Deep Investigation workflow builds a persistent research vault with landscape analysis, scored entity catalog, and cross-linked entity profiles. ## Quick Start Ask the assistant to do standard research on a topic such as recent developments in small language models.

Frequently Asked Questions about Research

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

FAQPage Schema
How do I run multi-agent web research with Claude?▼

Say 'research' followed by your topic to trigger Standard mode, which launches Claude and Gemini researcher agents in parallel. Say 'extensive research' for nine agents across Claude, Gemini, and Grok, or 'quick research' for a single fast lookup.

What is the difference between quick, standard, and extensive research modes?▼

Quick mode uses one agent and one query for fast answers in about 10-15 seconds. Standard mode runs two agents with one query each, while Extensive mode launches nine parallel agents across three researcher types for comprehensive coverage.

How does the skill prevent hallucinated URLs in research results?▼

Every URL returned by research agents must pass the URL Verification Protocol before delivery. Each link is fetched to confirm a 200 status and that the content actually supports the citation, with failed links replaced or dropped.

Can it scrape sites protected by CAPTCHA or bot detection?▼

Yes, the Retrieve workflow escalates through three layers: built-in WebFetch first, then BrightData MCP for CAPTCHA and bot-detection bypass, and finally Apify MCP actors for specialized site extraction.

When should I use Deep Investigation instead of Extensive Research?▼

Use Deep Investigation for iterative research that builds a persistent vault, such as market mapping or threat landscape analysis. It discovers entities, scores them by value and effort, and deep-dives the highest-priority ones across multiple iterations.

Related Skills

Explore More Education & Research Skills→
reedling78reedling78

research

Gather sources and write a markdown research document to docs/research/.

Community
Intermediate
slantviewslantview
1

content-marketer

Generate SEO-optimized blog posts, social media updates, and email newsletters.

Community
Intermediate
larsboeslarsboes

Investigation

Compile OSINT data from public records and social media for investigations.

Community
Intermediate
NimbalystNimbalyst
25

deep-researcher

Classify queries and run parallel subagent research to generate cited reports.

Official
Advanced