research-agent

Automates recursive source-verified research with credibility assessment and structured output.

71|9|Updated Jan 12, 2026
One-click install
npx skills add https://github.com/liangdabiao/skill-ten-prompt-generator --skill research-agent
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research-agent
Source: https://github.com/liangdabiao/skill-ten-prompt-generator/tree/main/.claude/skills/research-agent
Command: npx skills add https://github.com/liangdabiao/skill-ten-prompt-generator --skill research-agent

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates deep research and source evaluation for comprehensive market and academic analysis, enabling users to extract valuable insights from vast information while maintaining provenance and credibility.

Core Features & Use Cases

  • Recursive Research Tree: Systematically decompose topics into core sub-questions to guide thorough investigation.
  • Source Tiering & Tracing: Prioritize primary sources and clearly label ancillary or anecdotal evidence with traceability.
  • Red Team Analysis: Provide contrasting viewpoints and robust critique to counteract confirmation bias.
  • Synthesis Matrix & Density Chain: Compare theories and outputs across multiple dimensions and densify insights into actionable conclusions.
  • Uncertainty Labeling: Explicitly indicate confidence levels and data gaps for informed decision-making.

Quick Start

  1. Provide a research topic or question to begin planning.
  2. Review the generated Research Tree and terminology, then approve to initiate searches.
  3. The Skill will perform recursive searches, evaluate sources, and return a structured set of insights with provenance and uncertainties.

Frequently Asked Questions about research-agent

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

FAQPage Schema
How do I conduct deep research with automated source evaluation and provenance tracing?

Deep research with source evaluation uses recursive planning to decompose topics into sub-questions, systematically prioritizing primary sources while clearly labeling ancillary evidence to ensure traceability and credibility.

What is red team analysis and how does it counteract confirmation bias in competitive analysis?

Red team analysis provides contrasting viewpoints and robust critique during competitive analysis, systematically challenging initial assumptions to counteract confirmation bias and strengthen final conclusions.

How do I start a comprehensive literature review using a recursive research tree?

To start a literature review, provide a research topic to generate a recursive research tree of sub-questions, approve the plan, and the system will execute searches while returning structured insights with uncertainty labeling.

Can I use this approach for market research that requires explicit confidence levels and data gap identification?

Yes, market research applies uncertainty labeling to explicitly indicate confidence levels and identify data gaps, enabling informed decision-making by synthesizing multi-dimensional comparisons into actionable conclusions.

What is the best way to compare theories across multiple dimensions during an academic review?

The best way to compare theories during an academic review uses a synthesis matrix to evaluate differences across multiple dimensions, densifying extracted insights into structured, actionable conclusions with full source provenance.

When should I not use automated recursive planning for industry analysis?

Avoid automated recursive planning for industry analysis when your task requires real-time data execution or lacks clear definable boundaries, as the system relies on systematically decomposing topics into structured sub-questions.