skill-2-agentic-concern-extraction

Extract structured agentic concern sheets from agentic review outputs.

1|Updated Apr 9, 2026
One-click install
npx skills add https://github.com/jinming99/reviewer-under-review --skill skill-2-agentic-concern-extraction
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-2-agentic-concern-extraction
Source: https://github.com/jinming99/reviewer-under-review/tree/main/.claude/skills/skill-2-agentic-concern-extraction
Command: npx skills add https://github.com/jinming99/reviewer-under-review --skill skill-2-agentic-concern-extraction

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill automates turning agentic review outputs into a single, structured concern sheet to streamline evidence synthesis and verdict justification.

Core Features & Use Cases

  • Structured extraction: pulls decisiveness, major/minor concerns, goals, and decision drivers from review outputs (summary.yaml, review.md, adversarial_brief.md, gates.md, scorecard.md) and preserves provenance.
  • Normalization & deduplication: normalizes severities, consolidates duplicates across sources, and records source details for traceability.
  • Output formatting: produces an AgenticConcernSheet compliant with the calibration schema for downstream analysis and auditing.
  • Use Case: use this when building concern-alignment data for a paper, ensuring consistent representation of concerns and verdict drivers across reviews.

Quick Start

Run the agentic concern extractor on a paper's result directory to produce a structured concern sheet.

Frequently Asked Questions about skill-2-agentic-concern-extraction

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

FAQPage Schema
How do I extract structured concerns from agentic review outputs?

To extract structured concerns from agentic review outputs, run the agentic concern extractor on a paper's result directory. It integrates major/minor concerns, verdict drivers, and provenance from summary.yaml, review.md, adversarial_brief.md, gates.md, and scorecard.md into a single report.

What is an agentic concern sheet and when do I need one?

An agentic concern sheet is a schema-compliant report that normalizes severities, deduplicates concerns, and records source details for traceability. You need one when building concern-alignment data for a paper to ensure consistent representation of concerns across reviews.

How do I normalize severities and deduplicate concerns across multiple review files?

Normalize severities and deduplicate concerns by running the extractor across summary.yaml, review.md, adversarial_brief.md, gates.md, and scorecard.md. It consolidates duplicate concerns across sources and produces a schema-compliant AgenticConcernSheet with full provenance for downstream analysis.

Can I use the agentic concern extractor on multiple paper versions or methods?

The agentic concern extractor applies to single paper, single method, and single version result directories. It is designed to integrate decisive drivers, major/minor concerns, and decision drivers into a single report for one specific review scope.

What file formats do I need to generate an AgenticConcernSheet?

Generating an AgenticConcernSheet requires agentic review outputs in YAML and Markdown formats, specifically summary.yaml, review.md, adversarial_brief.md, gates.md, and scorecard.md. These files provide the decisive drivers, concerns, and origins extracted into the final report.