skills.rest
dandaka
dandakaCOMMUNITY
Explore all skills by this author →
Website𝕏@dandakaGitHub

deep-research

Synthesize multi-source research into structured reports with inline citations.

6|2|Updated Feb 17, 2026
Search Tags:#research#verification#multi-source#reporting#citations#parallel-agents#adversarial-review
One-click install
npx skills add https://github.com/dandaka/skills --skill deep-research-dandaka
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/dandaka/skills/tree/main/deep-research
Command: npx skills add https://github.com/dandaka/skills --skill deep-research-dandaka

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Facilitate rigorous, source-backed research within Claude Code, delivering structured, citable results and reducing bias from single-source workflows.

Core Features & Use Cases

  • Orchestrates structured, multi-source inquiries with parallel subagents
  • Enforces inline citations, source quality tiers, contradictions analysis, and cross-source validation
  • Supports Standard mode (10+ sources) and Deep mode (20+ sources with adversarial review and UNION merge)
  • Generates a publish-ready report with a pre-publish checklist and audit trail

Quick Start

Provide a topic prompt to begin a standard research run and append deep for a multi-agent deep-dive with 20+ sources.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I do multi-source research with inline citations?▼

Multi-source research with inline citations is achieved by synthesizing information from at least ten diverse sources, applying cross-source validation, and delivering a structured report with source metadata.

What is the best way to verify claims from multiple sources?▼

Verifying claims from multiple sources is best handled by applying adversarial review with parallel subagents in deep mode, which cross-validates over 20 sources and explicitly discusses any contradictions or uncertainties.

When do I need deep mode for high-stakes research reporting?▼

Deep mode for high-stakes research reporting is needed when a topic requires 20 or more sources, parallel subagents, adversarial review, and a UNION merge to ensure rigorous cross-source validation.

Can I generate a publish-ready research report with an audit trail?▼

Generating a publish-ready research report with an audit trail is supported by delivering structured results, a pre-publish checklist, and explicit source metadata to reduce bias from single-source workflows.

How does cross-source validation handle contradictions in research?▼

Cross-source validation handles contradictions in research by applying adversarial review and explicitly discussing any uncertainties or contradictions found across the diverse sources in the final report.

Related Skills

Explore More Education & Research Skills→
gopforevergopforever

deep-research

Coordinate parallel subagents to research topics and generate cited reports.

Community
Advanced
ArieGoldkinArieGoldkin
10

Quality Gates Skill

Review task assignments and enforce quality gates before work begins.

Community
Advanced
diShine-digital-agencydiShine-digital-agency
1

shine-web-research

Conduct multi-source web research with tiered search and citations.

Community
Advanced
Zidane786Zidane786

web-research

Plan, execute, and synthesize web research with parallel subagents into a cited report.

Community
Intermediate