skills.rest
CorellisOrg
CorellisOrgCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

deep-research

Route research questions across OpenAI, Gemini, and Claude models to generate cited reports.

27|1|Updated Mar 19, 2026
Search Tags:#synthesis#structured-output#deep-research#parallel-search#report-generation#multi-model-routing
One-click install
npx skills add https://github.com/CorellisOrg/Corellis --skill deep-research-corellisorg
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/CorellisOrg/Corellis/tree/main/templates/skills/deep-research
Command: npx skills add https://github.com/CorellisOrg/Corellis --skill deep-research-corellisorg

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires jq, curl, python3, and includes scripts (resource) components.

What problem does it solve?

Automates multi-model deep research routing and synthesis across OpenAI, Gemini, Claude, and Google Deep Research to produce comprehensive, sourced reports. It enables researchers to quickly obtain cross-validated insights by routing to the most suitable model and by coordinating parallel searches and deep fetching of sources.

Core Features & Use Cases

  • Multi-model routing: route questions to OpenAI, Gemini, Claude, or Google Deep Research and select the best model for the task.
  • Parallel search and deep fetch: run sub-questions in parallel, fetch full texts, and extract key insights with citations.
  • Synthesis & reporting: generate structured executive summaries, findings, and source-backed conclusions in multiple formats (canvas-ready, PDFs, or plain reports).
  • Asynchronous DR workflow: support long-running deep-research tasks with sub-agent orchestration and progress monitoring.

Quick Start

Ask it to perform deep research on a topic using the default routing and generate a comprehensive report.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I automate multi-model deep research routing to generate sourced reports?▼

You can automate deep research by routing sub-questions to OpenAI, Gemini, Claude, or Google Deep Research, running parallel searches to fetch full texts and synthesize structured sourced reports.

What is the best way to synthesize cross-validated insights from multiple AI models?▼

The best way to synthesize cross-validated insights is routing questions to multiple models like OpenAI and Claude, fetching sources in parallel, and generating source-backed structured conclusions.

How does asynchronous deep research workflow handle long-running search tasks?▼

Asynchronous deep research workflow handles long-running tasks by orchestrating sub-agents and monitoring progress, enabling deep fetching of sources without timing out during report generation.

Do I need python3 and curl installed to run parallel search and deep fetch operations?▼

Yes, you need python3, curl, and jq installed as dependencies to execute parallel search operations, route queries across different models, and process asynchronous deep research tasks.

Can I generate structured executive summaries and PDFs from multi-model research outputs?▼

Yes, multi-model research synthesis supports generating structured executive summaries and source-backed conclusions in multiple formats including canvas-ready reports, PDFs, and plain text outputs.

How do I route complex research topics with rich sub-questions to the most suitable AI model?▼

Model routing evaluates complex research topics with rich sub-questions and assigns each query to the most suitable AI model among OpenAI, Gemini, Claude, or Google Deep Research for optimal synthesis.

Related Skills

Explore More Education & Research Skills→
bowen31337bowen31337

weex-trading

Execute trades and manage positions on the WEEX exchange.

Community
Advanced
brainbrewlabsbrainbrewlabs
7

report-writing

Convert synthesis documents into structured Markdown research reports with APA citations.

Official
Intermediate
adeboldadebold

swarm-advanced

Coordinate swarm orchestration across mesh, hierarchical, star, and ring topologies for research, development, and testing workflows.

Community
Advanced
ThrookieThrookie
2

deep-research

Consolidate multi-source searches into a cited research report.

Community
Advanced