deep-research

Generate format-controlled research reports with evidence mapping and citation tracking.

35|2|Updated Feb 2, 2026
One-click install
npx skills add https://github.com/proyecto26/sherlock-ai-plugin --skill deep-research-proyecto26
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/proyecto26/sherlock-ai-plugin/tree/main/skills/deep-research
Command: npx skills add https://github.com/proyecto26/sherlock-ai-plugin --skill deep-research-proyecto26

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill enables users to generate high-fidelity, format-controlled research reports with built-in evidence tracking and citation management, reducing formatting drift and ensuring traceability.

Core Features & Use Cases

  • Evidence mapping: Create a traceable mapping from findings to sources.
  • Multi-pass drafting: Run parallel subagents to draft different report sections and synthesize into a single document.
  • UNION merge & format compliance: Merge multiple drafts while preserving structure and citations per the template.
  • Iterative review workflow: Present drafts for human review and incorporate feedback before finalization.

Quick Start

  1. Define the report spec (audience, scope, format, citation style).
  2. Build an outline and evidence table, then run multi-pass drafting.
  3. Perform the UNION merge, verify citations, and prepare the final report for delivery.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I generate a research report with traceable citations and evidence mapping?

To generate a research report with traceable citations, define your report spec, build an evidence table, and use multi-pass drafting to synthesize findings. The system maps findings directly to sources, ensuring traceability and reducing formatting drift.

What is multi-pass drafting for literature reviews and market analyses?

Multi-pass drafting runs parallel subagents to draft different report sections simultaneously, then merges them into a single document. This iterative synthesis ensures strict template adherence and comprehensive evidence collection before finalization.

Can I use this for policy briefs and technical briefs requiring strict formatting?

Yes, this workflow applies directly to policy briefs and technical briefs requiring strict formatting. It enforces template adherence and format compliance through a structured intake, evidence collection, drafting, and UNION merge process.

How do I maintain citation tracking and format compliance when merging multiple drafts?

You maintain citation tracking during a UNION merge, which combines multiple drafts while preserving the original structure and citations per your template. This ensures format compliance and traceability across the synthesized document.

Does the research report workflow support human review and feedback incorporation?

Yes, the iterative review workflow presents drafts for human review and incorporates feedback before finalization. This ensures the final report meets audience expectations and scope requirements while maintaining evidence mapping integrity.

What is the best way to structure a research report workflow for high-fidelity output?

The best way to structure a research report workflow is defining the report spec, building an outline and evidence table, running multi-pass drafting, and performing a UNION merge. This reduces formatting drift and ensures high-fidelity output.