deep-research

Synthesize multi-source research into structured reports with citation tracking.

4|Updated Feb 20, 2023
One-click install
npx skills add https://github.com/Hayao0819/dotfiles --skill deep-research-hayao0819
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/Hayao0819/dotfiles/tree/main/modules/home-manager/llm/skills/deep-research
Command: npx skills add https://github.com/Hayao0819/dotfiles --skill deep-research-hayao0819

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Multi-source research is often manual, opaque, and hard to audit. This Skill implements a citation-tracked, end-to-end research pipeline that persistently stores evidence and maintains context across sections, enabling reproducible findings and robust verification.

Core Features & Use Cases

  • Citation-tracked provenance: Every claim is tied to stable sources and an evidence corpus to enable auditability.
  • Evidence persistence: Quotes, data points, and source metadata are stored in append-only artifacts for traceable review.
  • Structured reporting: Produces a complete, narrative, yet rigorous report with executive summaries, findings, synthesis, and a complete bibliography.
  • Quality gates and verification: Built-in quality checks guard against placeholders, incomplete citations, and misaligned claims.
  • Use cases: Ideal for technology state-of-the-art reviews, competitive analyses, and compliance-ready research briefs where traceability matters.

Quick Start

Provide a research question to start a citation-tracked, multi-source analysis that retrieves sources, persists evidence, and outputs a structured 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 tracking?

You can automate multi-source research by providing a research question to a deterministic pipeline that retrieves sources, persists evidence, and outputs a structured report with stable citations. This process maintains an append-only evidence corpus for traceable review.

How does evidence persistence work in structured report generation?

Evidence persistence works by storing quotes, data points, and source metadata in append-only artifacts during structured report generation. This ensures every claim is tied to an evidence corpus, enabling reproducible findings and robust verification.

Can I use this for compliance-ready research briefs where traceability matters?

Yes, this is ideal for compliance-ready research briefs where traceability matters. It applies quality gates to guard against incomplete citations and maintains provenance tracking for each claim, source, and piece of evidence.

What is the best way to validate claims and manage context across report sections?

The best way to validate claims and manage context across report sections is using an autonomous pipeline that applies deterministic quality checks. It validates claims against an evidence corpus and maintains context throughout the synthesis process.

Does this research pipeline handle error handling and provenance tracking autonomously?

Yes, the research pipeline operates autonomously with explicit error handling and provenance tracking. It manages context, validates claims, and produces a complete narrative report with executive summaries, findings, and a bibliography.