deep-research

Orchestrate multi-source retrieval, triangulation, synthesis, and critique into validated reports.

1|1|Updated Feb 13, 2026
One-click install
npx skills add https://github.com/eskoNBG/UltiDB --skill deep-research-eskonbg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/eskoNBG/UltiDB/tree/main/skills/deep-research
Command: npx skills add https://github.com/eskoNBG/UltiDB --skill deep-research-eskonbg

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Automates enterprise-grade research workflows by orchestrating multi-source retrieval, validation, and synthesis to produce credible, citation-backed reports without requiring manual curation.

Core Features & Use Cases

  • Multi-source retrieval: parallel gathering from diverse sources with credibility scoring and source tracking.
  • 3+ source triangulation: cross-checks claims across at least three independent sources.
  • Automated validation: eight checks including executive summary length, citation formatting, bibliography completeness, and consistency checks.
  • Synthesis & critique: red-team style critique and iterative refinement to improve quality before packaging.
  • Output formats: Markdown, HTML (McKinsey style), and PDF deliverables with structured presentation.
  • Use case: supports complex decisions requiring 10+ sources, verified claims, and cross-domain validation.

Quick Start

Invoke the GLM-based research engine in standard mode with a complex query to generate a comprehensive, citation-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 credibility scoring?

Automate multi-source research by orchestrating parallel retrieval from diverse sources, applying credibility scoring, and triangulating claims across at least three independent sources to generate citation-backed reports.

What is source triangulation and when do I need it for complex analytical questions?

Source triangulation cross-checks claims across three or more independent sources. It is needed for complex analytical questions requiring 10+ sources, high credibility, and cross-source verification to ensure accuracy.

How do I validate research reports for citation formatting and bibliography completeness?

Validate research reports using eight automated checks that verify executive summary length, citation formatting, bibliography completeness, and consistency to ensure structured output quality before packaging.

Can I generate McKinsey style HTML and PDF deliverables from markdown research reports?

Generate McKinsey style HTML and PDF deliverables from markdown by applying synthesis, red-team critique, and iterative refinement to produce structured presentation outputs suitable for enterprise use.

Does automated deep research support self-contained operation using only Python stdlib?

Automated deep research supports self-contained operation using only Python stdlib, with optional extensions available for web and document processing, enabling local execution without external dependencies.

What are the limitations of automated validation in deep research pipelines?

Automated validation is limited to eight specific checks including executive summary length, citation formatting, and bibliography completeness, focusing on structural consistency rather than evaluating subjective analytical quality.