skills.rest
YSheldon
YSheldonCOMMUNITY
Explore all skills by this author →
GitHub

deep-research

Orchestrate parallel subagents to research topics and synthesize cited reports.

Updated Apr 12, 2026
Search Tags:#research#subagents#citations#synthesis#web-search#phase-planning#multisource
One-click install
npx skills add https://github.com/YSheldon/Prompt-Log --skill deep-research-ysheldon
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/YSheldon/Prompt-Log/tree/main/.local/secondary_skills/deep-research
Command: npx skills add https://github.com/YSheldon/Prompt-Log --skill deep-research-ysheldon

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Complex topics demand exhaustive, multi-source investigation and structured, cite-worthy outputs; this skill reduces manual research overhead by orchestrating source gathering, evaluation, and synthesis.

Core Features & Use Cases

  • Parallel focus areas: Decomposes topics into 5 non-overlapping angles and runs subagents in parallel to accelerate discovery.
  • Structured synthesis: Organizes findings into a report with citations, gaps, and cross-source analysis.
  • Governance & Quality: Includes source evaluation for credibility and timestamps research outputs.
  • Use Case: Ideal for literature reviews, market analyses, technology evaluations, and risk assessments requiring cross-source validation.

Quick Start

Initiate a deep, multi-source research session by defining scope, launching parallel subagents, and producing a cited synthesis.

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-source research and synthesis with citations?▼

Multi-source research and synthesis is automated by decomposing topics into non-overlapping angles, running parallel subagents for web-search discovery, and organizing findings into a published report with citations and cross-source analysis.

What is the best way to conduct a literature review across multiple web sources?▼

A literature review across multiple web sources is best conducted by applying phase-planning to evaluate source credibility, executing parallel subagents for discovery, and synthesizing cross-validated findings into a structured report.

Can I use parallel subagents to accelerate technology assessments and market analyses?▼

Parallel subagents can be used to accelerate technology assessments and market analyses by simultaneously gathering and evaluating multi-source data, reducing manual research overhead while producing timestamped outputs.

How does cross-source validation work during structured research?▼

Cross-source validation works by evaluating gathered web-search data for credibility, identifying information gaps, and synthesizing findings across multiple sources to ensure the final report is thorough and cite-worthy.

When do I need phase-planning for a deep research session?▼

Phase-planning is needed for a deep research session when complex topics demand exhaustive, multi-source investigation, requiring structured decomposition into distinct angles before running parallel subagents.

Are there limitations to using automated subagents for research synthesis?▼

Automated subagents for research synthesis are limited by web-search source availability and credibility; while source evaluation is applied, outputs require review to ensure cross-source gaps are fully addressed.

Related Skills

Explore More Education & Research Skills→
LayerDynamicsLayerDynamics

dispatching-parallel-agents

Dispatch parallel agents with self-contained prompts for concurrent independent tasks.

Community
Advanced
drillandrillan

research

Automate sourced, cited research with structured JSON outputs via the Tavily CLI script.

Community
Advanced
DevHive1DevHive1
18

deep-research

Decompose complex topics into parallel research areas and synthesize cited reports.

Community
Advanced
ItsJupiter000ItsJupiter000

deep-research

Coordinate multi-source research into structured, cited findings.

Community
Advanced