skills.rest
ShinyGua
ShinyGuaCOMMUNITY
Explore all skills by this author →
GitHub

ar-evidence

Extract evidence units from reading notes and group them by theme and sub-question.

Updated Mar 22, 2026
Search Tags:#evidence#nlp#source-traceability#reading-notes#contradiction-detection#thematic-grouping#evidence-index
One-click install
npx skills add https://github.com/ShinyGua/AutoArtsResearch --skill ar-evidence
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ar-evidence
Source: https://github.com/ShinyGua/AutoArtsResearch/tree/main/.claude/skills/ar-evidence
Command: npx skills add https://github.com/ShinyGua/AutoArtsResearch --skill ar-evidence

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Transform reading notes into a structured set of evidence units, enabling clear traceability from source to insight and supporting robust argument construction.

Core Features & Use Cases

  • Extracts evidence units (quotes, facts, statistics, interpretations) from reading notes and records metadata such as source, span, type, and confidence.
  • Groups evidence by sub-question and by literature-map themes to build a navigable evidence index.
  • Identifies and records contradictions between evidence units to surface competing claims and uncertainties.
  • Writes an evidence index and individual evidence unit records to the workspace to support downstream claims and peer-review workflows.

Quick Start

Provide a workspace path containing reading notes and run the ar-evidence skill to construct a structured evidence index.

Frequently Asked Questions about ar-evidence

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

FAQPage Schema
How does contradiction detection work when organizing research evidence?▼

Source traceability in research is maintained by extracting evidence units from reading notes and recording metadata such as source, span, type, and confidence for each unit in an individual JSON record.

What output format does evidence extraction generate for research workflows?▼

Extracted reading notes are written as structured output to a workspace path, generating an evidence index at analysis/evidence/evidence_index.json alongside individual ev-*.json files for each unit.

Do I need a specific workspace path to run evidence extraction on reading notes?▼

Evidence extraction generates an evidence_index.json file at {workspace}/analysis/evidence/ alongside individual ev-*.json files, containing grouped evidence units and recorded contradictions to support peer-review workflows.

What metadata is recorded for evidence units extracted from reading notes?▼

Evidence extraction records metadata including source, span, type, and confidence for each evidence unit, ensuring complete source traceability from original reading notes to the final structured evidence index.

Does source traceability work for literature reviews using reading notes?▼

Evidence extraction records metadata including source, span, type, and confidence for each evidence unit, ensuring complete source traceability from original reading notes to the final structured evidence index.

Related Skills

Explore More Education & Research Skills→
onebrain-aionebrain-ai
23

reading-notes

Convert finished books or articles into structured progressive summary notes.

Official
Intermediate
deerawandeerawan
1

gsd-review

Coordinate cross-AI peer reviews of phase planning documents and write consolidated REVIEWS.md output.

Community
Intermediate
theamtheam
36

exploratory-sota-research

Map AI/ML state-of-the-art research into mechanism-based landscapes with evidence synthesis.

Official
Advanced
VitalPointAIVitalPointAI

NEAR Transaction Labeler

Categorize NEAR transactions and export labeled data as CSV or JSON.

Official
Intermediate