deep-research

Coordinate multi-agent research to decompose questions and synthesize cited findings.

Updated Apr 15, 2026

Search Tags:#planning #research #verification #multi-agent #synthesis #guardrails #domain-research

Ready to run

npx skills add https://github.com/jasonmichaelbell78-creator/JASON-OS --skill deep-research-jasonmichaelbell78-creator

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/jasonmichaelbell78-creator/JASON-OS/tree/main/.claude/skills/deep-research
Command: npx skills add https://github.com/jasonmichaelbell78-creator/JASON-OS --skill deep-research-jasonmichaelbell78-creator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the challenge of answering complex questions by coordinating multiple researchers (agents) to decompose problems, execute parallel searches, synthesize findings with citations and confidence levels, and route outputs through downstream adapters with guardrails.

Core Features & Use Cases

Decompose questions into MECE sub-questions and allocate agents accordingly.
Dispatch parallel searcher agents to gather diverse sources and track confidence.
Synthesize findings with citations, confidence, and self-audit checks; include contrarian/OTB challenges.
Route outputs to downstream adapters and planning tools for skill creation and roadmapping.
Maintain resumable state and versioned outputs for traceability.

Quick Start

Invoke the deep-research skill with a topic to start a structured, multi-agent investigation.

Frequently Asked Questions about deep-research

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

FAQPage Schema

How do I decompose complex research questions into sub-questions for parallel search?▼

Multi-agent research coordinates parallel searcher agents to execute diverse source gathering, structuring findings with citations, confidence levels, and self-audit checks across all phases.

How does multi-agent research handle synthesis and verification?▼

Multi-agent research coordinates parallel searcher agents to execute diverse source gathering, structuring findings with citations, confidence levels, and self-audit checks across all phases.

What is the best way to add contrarian and out-of-the-box challenges to research synthesis?▼

Enforce research guardrails by applying plan-first execution, maintaining floor depth requirements, and running mandatory verification checks across all investigation phases to ensure traceability.

How do I enforce guardrails like plan-first execution and mandatory verification in research?▼

Enforce research guardrails by applying plan-first execution, maintaining floor depth requirements, and running mandatory verification checks across all investigation phases to ensure traceability.

Can I route multi-agent research outputs to downstream adapters for skill creation and roadmapping?▼

Route multi-agent research outputs to downstream adapters and planning tools to generate versioned outputs, maintain resumable state, and support skill creation and roadmapping tasks.

What are the limitations of using automated multi-agent research for domain investigations?▼

Limitations of multi-agent research include dependency on plan-first execution constraints, floor depth enforcement, and mandatory verification, which may restrict unstructured domain investigations or exploratory research.