figure-first-paper-reader

Reads research papers figure by figure to audit whether visuals support the authors' claims.

Updated Sep 15, 2026
One-click install
npx skills add https://github.com/mrsonord2240/openscience-specialists --skill figure-first-paper-reader-mrsonord2240
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: figure-first-paper-reader
Source: https://github.com/mrsonord2240/openscience-specialists/tree/main/specialists/critical-appraisal-specialist/versions/1.0.0/package/skills/figure-first-paper-reader
Command: npx skills add https://github.com/mrsonord2240/openscience-specialists --skill figure-first-paper-reader-mrsonord2240

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Reading a full biomedical paper is slow, and standard summaries often repeat the authors' narrative without checking whether the figures actually support the conclusions. This Skill reconstructs a paper's logic figure by figure and judges the evidentiary strength of each visual, so you can quickly see which claims hold up and where the narrative overreaches. ## Core Features & Use Cases - Figure-to-Claim Mapping: Separates what each figure visibly shows from the claim the authors attach to it, classifying evidence as descriptive, comparative, mechanistic, predictive, or validation-oriented. - Support Strength Judgment: Rates each figure as strong, partial, weak, or insufficient support, and flags overinterpretation such as turning associations into causal claims. - Narrative Reconstruction: Rebuilds the paper's argument in figure order, identifies the 1-3 core figures carrying the main claims, and delivers a structured report with a self-critical risk review. - Use Case: Before a journal club, provide a paper's figures and captions and ask for a figure-first read to learn which visuals genuinely support the headline conclusion and where the discussion overstates the evidence. ## Quick Start Read this paper figure by figure and tell me whether the visuals actually support the main claims.

Frequently Asked Questions about figure-first-paper-reader

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

FAQPage Schema
How do I quickly understand a research paper without reading every paragraph?▼

Use a figure-first reading approach that extracts the paper's logic from its figures in order. Each figure is mapped to the claim it supports, letting you grasp the core findings and argument structure before deciding whether a full read is worthwhile.

How can I check whether a paper's figures actually support its conclusions?▼

Provide the figures, captions, or paper text and request a figure-to-claim audit. Each figure is classified as strong, partial, weak, or insufficient support, and overinterpretation such as turning associations into causal claims is explicitly flagged.

Can this skill read a paper if I only have figure captions or screenshots?▼

Yes, it accepts full figures, screenshots, captions only, or paper text describing the figures. When material is incomplete, cropped, or unreadable, the read is explicitly labeled as limited or provisional rather than overclaiming certainty.

Does a figure-first read replace full critical appraisal of a study?▼

No. A figure-first read tests whether visuals support the claims but does not evaluate methods, sample construction, or statistical handling. For those concerns, follow up with a study-design or methods appraisal skill.

What are the limitations of figure-based paper reading?▼

It cannot infer hidden numeric values, p-values, or sample sizes not visible in the figures, and it will not invent unseen panel contents. Missing or illegible figures are flagged as information gaps rather than filled in.