/research

Decompose complex research queries into parallel agent workstreams for cited markdown reports.

18|Updated Mar 6, 2026
One-click install
npx skills add https://github.com/altmbr/claude-research-skill --skill research-altmbr
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: /research
Source: https://github.com/altmbr/claude-research-skill/tree/main
Command: npx skills add https://github.com/altmbr/claude-research-skill --skill research-altmbr

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the bottleneck of manual research by automating the decomposition, execution, and synthesis of complex questions into a single, high-quality document.

Core Features & Use Cases

  • Parallel Workstreams: Automatically breaks down broad questions into 2-4 specialized agent tasks that run simultaneously.
  • Stuck Agent Recovery: Monitors progress in real-time and automatically relaunches agents that hit roadblocks or loops.
  • Unified Synthesis: Aggregates disparate research findings into a coherent report with executive summaries, cross-cutting themes, and confidence assessments.

Quick Start

Type /research followed by your complex question to initiate the multi-agent research workflow.

Frequently Asked Questions about /research

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

FAQPage Schema
How do I automate research and synthesize findings into a cited report?

Multi-agent research automation decomposes complex queries into parallel workstreams, searches for information, and synthesizes findings into a single cited markdown report. Specialized agents run simultaneously to gather deep information across diverse topics.

Can I run parallel research agents to cover multiple topics at once?

Yes, parallel workstreams break broad questions into 2-4 specialized agent tasks that run simultaneously. This orchestration accelerates research by executing multiple deep-dive information gathering streams at the same time.

What is the best way to generate a comprehensive report from disparate research sources?

Unified synthesis aggregates disparate research findings into a coherent markdown report featuring executive summaries, cross-cutting themes, and confidence assessments to ensure source integrity and comprehensive coverage.

Do I need Claude Code's Task tool to manage multi-agent research orchestration?

Yes, the research orchestration requires Claude Code's Task tool to manage background subagents and maintain strict write-after-search protocols during the automated report generation workflow.

How are stuck research agents recovered during automated analysis?

The orchestration system monitors agent progress in real-time and automatically relaunches any research agents that hit roadblocks or enter loops, ensuring workstreams complete successfully.