deep-research

Retrieve and validate multi-source evidence to build citation-backed research reports.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/lightxue/dotfiles --skill deep-research-lightxue
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/lightxue/dotfiles/tree/main/agent/skills/deep-research
Command: npx skills add https://github.com/lightxue/dotfiles --skill deep-research-lightxue

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Enterprise teams often need thorough, verifiable research that aggregates multiple sources, assesses credibility, and preserves provenance. This skill automates the end-to-end process, delivering citation-backed reports with structured pipelines and traceable evidence.

Core Features & Use Cases

  • Multi-source retrieval: pulls information from web, academic, and professional sources with parallel processing.
  • Credibility scoring and verification: assigns credibility scores, flags potential biases, and performs DOI/URL validation.
  • Progressive report packaging: builds executive summaries, findings, synthesis, and a complete bibliography with a machine-readable provenance trail.
  • Use Case: generate a comprehensive technical analysis comparing competing architectures, with source-backed conclusions and actionable recommendations.

Quick Start

Trigger the skill with a research topic to start automatic, multi-source reporting.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I automatically generate a citation-backed research report from multiple sources?

Automatically generate a citation-backed research report by triggering the skill with a topic to initiate multi-source retrieval, credibility scoring, and progressive packaging into a structured document with a bibliography.

What is the best way to verify evidence and score source credibility for a technical analysis?

The best way to verify evidence and score source credibility is through automated validation that flags potential biases and performs DOI/URL validation across academic and technical sources during aggregation.

Can I compile research findings from web and academic sources with traceable provenance?

Yes, you can compile research findings with traceable provenance by applying parallel multi-source retrieval across web and academic platforms, producing a machine-readable trail linking conclusions to their original sources.

How do I structure an executive summary and synthesis from aggregated research data?

Structure an executive summary and synthesis by using progressive report packaging, which automatically builds these sections alongside findings and a complete bibliography as the evidence base is assembled.

Does multi-source retrieval preserve context for continuation if I need to pause research?

Yes, multi-source retrieval preserves context for continuation, allowing you to pause and resume the automated compilation of verifiable citations and structured evidence without losing the gathered provenance trail.

What are the limitations of automated credibility scoring for verifiable citations?

Automated credibility scoring for verifiable citations relies on DOI/URL validation and bias flagging, meaning it may not fully capture nuanced contextual biases or evaluate sources behind inaccessible paywalls or non-standardized identifiers.