deep-research-main

Orchestrate multi-agent deep research workflows with phase contracts and inline citations.

Updated Mar 21, 2026
One-click install
npx skills add https://github.com/gtpgg1013/claude-skills-collection --skill deep-research-main
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research-main
Source: https://github.com/gtpgg1013/claude-skills-collection/tree/main/skills/plugins/gptaku/deep-research/skills/deep-research-main
Command: npx skills add https://github.com/gtpgg1013/claude-skills-collection --skill deep-research-main

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Deep Research Skill orchestrates a repeatable, multi-agent workflow to perform rigorous topic research, manage sources, verify claims, and produce structured outputs with inline citations.

Core Features & Use Cases

  • Phase-driven research pipeline: Scopes questions, plans retrieval, executes iterative queries, triangulates sources, synthesizes findings, and formats final deliverables.
  • Multi-agent collaboration: Deploys specialized agents (web, academic, verification) to maximize coverage, accuracy, and validation.
  • Stateful session management: Tracks progress, persists sources and artifacts, and supports resume and auditing.
  • Use Case: Teams researching emerging technologies can generate a full, citation-rich report with an executive summary and bibliography.

Quick Start

Provide a topic to initiate an end-to-end seven-phase deep-research workflow.

Frequently Asked Questions about deep-research-main

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

FAQPage Schema
How do I generate a citation-rich research report using a multi-agent workflow?

Generate a citation-rich research report by inputting a topic to initiate a multi-agent deep research workflow, which decomposes the subject into subtopics and assigns specialized agents to synthesize evidence-backed findings with inline citations.

What is the best way to manage state and track provenance across deep research phases?

Manage state and track provenance across deep research phases using explicit stateful session management, which persists sources, tracks progress, enforces phase contracts, and supports resume and auditing throughout the pipeline.

Can I use multi-agent collaboration to triangulate sources and verify claims for academic research?

Yes, multi-agent collaboration deploys specialized agents to maximize coverage, triangulate sources, verify claims, and validate findings, making it suitable for rigorous academic research and structured synthesis.

How do I scope questions and plan retrieval for an end-to-end deep research pipeline?

Scope questions and plan retrieval by initiating the end-to-end deep research workflow, which progresses through seven phases including scoping, planning, querying, triangulation, synthesis, QA, and packaging.

What are the limitations of using a phase-based workflow for deep research?

A phase-based workflow for deep research requires strict adherence to phase contracts, data provenance rules, and quality assurance rubrics, meaning skipping phases or unstructured inputs will break the stateful session management.