deep-research

Investigate complex questions and generate source-backed Markdown research reports.

313|131|Updated Mar 31, 2026
One-click install
npx skills add https://github.com/7df-lab/devo --skill deep-research-7df-lab
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/7df-lab/devo/tree/main/crates/skills/src/assets/samples/deep-research
Command: npx skills add https://github.com/7df-lab/devo --skill deep-research-7df-lab

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill replaces shallow, ad hoc searching with a rigorous research workflow that clarifies the question, gathers evidence, evaluates source quality, and produces a well-supported synthesis.

Core Features & Use Cases

  • Staged Research Workflow: Progress from clarification and research briefing through supervised investigation, evidence compression, and final report writing.
  • Source-Backed Synthesis: Coordinate web, fetch, code, and local inspection tools while preserving citations, provenance, conflicts, and uncertainty.
  • Structured Reporting: Create a substantial Markdown report with an executive summary, methodology, findings, analysis, limitations, conclusion, and references.
  • Use Case: Compare competing technologies, investigate a policy question, or analyze a market topic using multiple perspectives and clearly linked evidence.

Quick Start

Use the deep-research skill to investigate the user's question, verify claims with authoritative sources, and save a cited Markdown report.

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 cited report from a complex research question?

To generate a cited report, you investigate the complex question through a staged workflow involving clarification, agent-based evidence gathering, and source verification to produce a structured Markdown report with preserved citations.

What is the best way to perform evidence synthesis and comparative analysis on a policy question?

Evidence synthesis and comparative analysis require coordinating web search and fetch tools to gather multiple perspectives, track uncertainty, and compress findings into a well-supported synthesis laid out in a structured Markdown report.

Can I use deep research to investigate a market topic and track conflicting sources?

Yes, deep research investigates market topics by applying agent-based evidence gathering, preserving source provenance, tracking conflicts and uncertainty, and synthesizing the verified evidence into a cited report.

How does source verification work when gathering evidence for multi-perspective investigation?

Source verification works by evaluating source quality during supervised investigation, preserving citations and provenance, and tracking uncertainty to ensure the final evidence synthesis is rigorously source-backed.

Does this research workflow require staged clarification before writing the final Markdown report?

Yes, the research workflow requires staged clarification to define the research briefing, followed by supervised investigation and evidence compression, before generating the final Markdown report with an executive summary and references.

What are the limitations of using ad hoc searching for evidence synthesis instead of a structured research workflow?

Ad hoc searching lacks staged clarification, source verification, and uncertainty tracking, which limits its ability to produce rigorous, source-backed evidence synthesis with preserved citations and structured reporting.