deep-research

Execute an automated 8-phase research pipeline with source credibility scoring.

Updated Mar 30, 2026
One-click install
npx skills add https://github.com/erlebach/gordon --skill deep-research-erlebach
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/erlebach/gordon/tree/main/skills/research_skills/deep-research
Command: npx skills add https://github.com/erlebach/gordon --skill deep-research-erlebach

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Multi-source research for complex topics is time-consuming and error-prone; this Skill automates a rigorous, citation-backed analysis pipeline to deliver verified reports.

Core Features & Use Cases

  • Automated 8-phase research pipeline (Scope, Plan, Retrieve, Triangulate, Synthesize, Critique, Refine, Package) with source credibility scoring
  • Automated validation and conditional gating to ensure quality before delivery
  • 4 depth modes (quick, standard, deep, ultradeep) for speed-precision trade-offs
  • Local, self-contained operation with offline-friendly architecture (Claude Code skills)
  • Progressive context management and on-demand references for token efficiency

Quick Start

Ask Claude Code to run deep-research on your topic to generate a comprehensive, source-backed report.

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 citation management and source credibility scoring?

Automated multi-source research executes an 8-phase pipeline including retrieval, triangulation, and synthesis to generate verified reports. It applies source credibility scoring and conditional validation to ensure high-quality, citation-backed analysis across academic and technical domains.

What is the best way to synthesize complex topics requiring ten or more sources?

Synthesizing complex topics across 10+ sources requires cross-source triangulation and automated critique phases. An autonomous research pipeline retrieves, validates, and cross-references multiple sources to produce a self-contained report with managed context and on-demand references.

Can I run offline, self-contained research tasks without external interactive prompts?

Yes, offline self-contained research tasks operate without external interactive prompts. The system uses local storage for reports and progressive context management, executing the full research pipeline autonomously within a Claude Code environment.

How do I balance speed and precision when validating research across multiple sources?

Balancing speed and precision during multi-source validation is achieved using four depth modes: quick, standard, deep, and ultradeep. These mode variants adjust the automated research pipeline's speed-precision trade-off based on the required level of source verification.

Does automated research validation support cross-source synthesis across different domains?

Automated research validation supports cross-source synthesis across academic, industry, and technical domains. The pipeline includes an automated critique and refine phase with conditional gating to ensure triangulated data meets quality standards before packaging the local report.

What are the limitations of autonomous research pipelines for complex topic analysis?

Autonomous research pipelines are limited by their self-contained architecture and lack of external interactive prompts. While they support 10+ sources and local storage, complex topic analysis depends entirely on the selected depth mode variant and the system's built-in source credibility scoring capabilities.