unit-analyst

Standardize analysis of research units with evidence-bound prepare-fill-verify workflows.

5|Updated Apr 13, 2026
One-click install
npx skills add https://github.com/caozx1110/ResearchLab --skill unit-analyst
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: unit-analyst
Source: https://github.com/caozx1110/ResearchLab/tree/main/skills/unit-analyst
Command: npx skills add https://github.com/caozx1110/ResearchLab --skill unit-analyst

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill solves the problem of inconsistent and unverified analysis of research materials by providing a standardized, evidence-first workflow for papers, repositories, datasets, and blogs.

Core Features & Use Cases

  • Evidence-Bound Analysis: Ensures every claim made by the AI is backed by a verbatim quote and locator from the source material.
  • Kind-Specific Routing: Automatically routes analysis tasks to specialized implementations for different content types like papers or code repositories.
  • Verification Gate: Enforces a strict prepare-fill-verify cycle that prevents unverified or hollow content from entering the knowledge base.

Quick Start

Use the unit-analyst skill to prepare a deep-read analysis for the paper unit identified by the provided ID.

Frequently Asked Questions about unit-analyst

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

FAQPage Schema
How do I verify research analysis claims against source material?▼

To verify research analysis claims, an evidence-bound workflow enforces a strict prepare-fill-verify cycle that requires every AI-generated claim to be backed by a verbatim quote and source locator, preventing unverified content from entering the knowledge base.

What is the best way to standardize analysis across papers, datasets, and repositories?▼

Standardizing analysis across diverse research units is achieved through kind-specific routing, which automatically directs papers, repositories, datasets, and blogs to specialized implementations for structured, consistent evidence extraction.

How does an agent-led filling workflow ensure research integrity?▼

An agent-led filling workflow ensures research integrity by coordinating structured content generation within a managed runtime environment, validating all generated analysis claims against source parse-caches before final verification.

Do I need a managed runtime environment to execute research analysis scripts?▼

Yes, a managed runtime environment is required to execute the implementation scripts and validate analysis claims against source parse-caches during the verification gate phase of the research workflow.

Why does my research unit analysis output contain unverified or hollow content?▼

Unverified or hollow content appears when the verification gate is bypassed, meaning the strict prepare-fill-verify cycle was not completed to validate claims against the source parse-caches.

Can I use evidence-bound analysis for different content types like blogs?▼

Yes, evidence-bound analysis supports different content types like blogs through kind-specific routing, which automatically matches each research unit to a specialized implementation for accurate standardized processing.