deep-research

Coordinate multi-source retrieval, validation, and synthesis into cited research reports.

152|32|Updated Dec 29, 2025
One-click install
npx skills add https://github.com/lintendo/Axhub-Make --skill deep-research-lintendo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/lintendo/Axhub-Make/tree/main/skills/third-party/deep-research
Command: npx skills add https://github.com/lintendo/Axhub-Make --skill deep-research-lintendo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill orchestrates enterprise-grade research workflows to produce citation-backed, verified analyses by coordinating multi-source retrieval, validation, and synthesis.

Core Features & Use Cases

  • 8-phase pipeline: Scope → Plan → Retrieve → Triangulate → Synthesize → Critique → Refine → Package, with automated validation and credibility scoring.
  • Source credibility & verification: Per-source credibility scoring (0-100) and automated citation validation to ensure reliability.
  • Autonomy & efficiency: Operates with minimal user prompts, announces mode, and loads context progressively to optimize latency.
  • Output formats & templates: Generates Markdown, HTML (McKinsey style), and PDF reports with structured bibliographies.

Quick Start

Run Standard mode to generate a comprehensive, citation-backed research report with 15-30 sources.

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 with cross-source verification?

Automate multi-source research using an 8-phase pipeline that handles retrieval, triangulation, and synthesis. It cross-references 15-30 sources per topic, applying automated credibility scoring to generate a cited, defensible analysis.

What is source credibility scoring and how does it validate research?

Source credibility scoring evaluates each retrieved reference on a 0-100 scale during the validation phase. This automated citation validation checks reliability across domains, ensuring the final synthesis only depends on verified, high-quality sources.

How do I generate cited research reports in multiple formats like HTML or PDF?

Generate cited research reports in Markdown, HTML, or PDF formats using the pipeline's packaging phase. It outputs structured bibliographies and applies McKinsey-style formatting to HTML reports for professional presentation.

Can I run autonomous research workflows with minimal input prompts?

Yes, you can run autonomous research workflows in Standard mode with minimal input prompts. The system announces its mode, progressively loads context, and optimizes latency to produce comprehensive reports autonomously.

Does multi-source retrieval work for complex topics requiring 10 or more sources?

Multi-source retrieval is specifically designed for complex topics requiring 10 or more sources. It coordinates cross-domain validation and rigorous triangulation to handle demanding research questions that single-source lookups cannot resolve.

What's the best way to structure a research pipeline for defensible analysis?

Structure defensible analysis using an 8-phase pipeline: Scope, Plan, Retrieve, Triangulate, Synthesize, Critique, Refine, and Package. This approach enforces automated validation and progressive refinement to ensure rigorous results.