analyze

Analyze sources or claims through descriptive, evaluative, and dialectical stages.

38|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/lhl/realitycheck --skill analyze-lhl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analyze
Source: https://github.com/lhl/realitycheck/tree/main/integrations/claude/skills/analyze
Command: npx skills add https://github.com/lhl/realitycheck --skill analyze-lhl

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a structured, three-stage framework to analyze sources and claims, enabling rigorous, audit-ready assessments without auto-registration into a knowledge base.

Core Features & Use Cases

  • Descriptive Analysis: neutrally summarize the source, identify key claims, and establish context.
  • Evaluative Analysis: assess evidence quality using the Reality Check hierarchy and detect contradictions.
  • Dialectical Analysis: present steelman and strongest counterarguments to synthesize a mature understanding.
  • Use Case: analysts who want a transparent, manual critique before committing to database registration.

Quick Start

Run the analyze skill on a target source/ID with the command: /analyze <url_or_source_id> and review the generated three-stage report for audit-ready claims.

Frequently Asked Questions about analyze

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

FAQPage Schema
How do I perform a structured source analysis for an audit-ready report?

Structured source analysis uses a three-stage Reality Check process to generate an audit-ready report. It descriptively summarizes sources, evaluates evidence quality, and synthesizes dialectical counterarguments.

What is the best way to evaluate evidence quality and detect contradictions in a claim?

Evaluating evidence quality applies the Reality Check hierarchy to assess sources and detect contradictions. This evaluative stage measures evidence strength and identifies internal inconsistencies within the provided claims.

How do I critique a dataset comprehensively without auto-registering it into a knowledge base?

To critique a dataset without auto-registration, perform a manual three-stage analysis. This process produces a structured critique and candidate claims for later registration, keeping the database untouched until manually committed.

Can I synthesize steelman arguments and counterarguments for a complex claim?

Yes, synthesizing steelman arguments and counterarguments is achieved through dialectical analysis. This third stage presents the strongest opposing views to synthesize a mature, comprehensive understanding of the claim.

Does this manual analysis framework require any specific database dependencies?

No, this manual analysis framework requires no specific database dependencies. It operates independently to produce structured reports and candidate claims, explicitly avoiding automatic database registration during the analysis process.