skills.rest
bfairkun
bfairkunCOMMUNITY
Explore all skills by this author →
GitHub

paper-reading

Retrieve and verify scientific paper full text via Europe PMC, PMC, and bioRxiv before summarizing.

1|Updated Oct 25, 2020
Search Tags:#literature review#scientific papers#hallucination prevention#biorxiv#doi#full text retrieval#pubmed central
One-click install
npx skills add https://github.com/bfairkun/dotfiles --skill paper-reading-bfairkun
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: paper-reading
Source: https://github.com/bfairkun/dotfiles/tree/main/agents/.agents/skills/paper-reading
Command: npx skills add https://github.com/bfairkun/dotfiles --skill paper-reading-bfairkun

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? AI assistants often describe papers from abstracts or search snippets, producing confident but wrong claims about methods and results. This Skill enforces retrieval of the real full text before any claim is made, and labels every claim by what was actually read. ## Core Features & Use Cases - Identifier resolution and full-text retrieval: Resolves DOI, PMID, PMCID, or title through Europe PMC and NCBI's ID converter, then fetches full text from Europe PMC XML, PMC HTML, or bioRxiv/medRxiv, printing a provenance block with source and character count. - Honest failure signaling: Exit codes distinguish full text (0), abstract only (2), and nothing reachable (3), so claims are labeled accurately instead of fabricated. - Figure image download: Pulls figure images from PMC's CDN with --figures for genuine visual inspection rather than relying on captions. - Use Case: Given a DOI for a paywalled Cell paper, run the retrieval script, discover the PMC deposit has the full text despite being non-open-access, and summarize the actual Methods section instead of guessing from the abstract. ## Quick Start Ask the AI to read the paper at DOI 10.xxxx/example using the paper-reading skill and summarize only what the full text actually says.

Frequently Asked Questions about paper-reading

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

FAQPage Schema
How do I get the full text of a paper from a DOI or PMID?▼

Run the get_fulltext.py script with the DOI, PMID, PMCID, or title words. It resolves the identifier through Europe PMC and NCBI's ID converter, then tries Europe PMC full-text XML, PMC HTML, and bioRxiv/medRxiv in order, printing a provenance block with the source.

How to read a paywalled paper without institutional VPN access?▼

Try PMC first, since many non-open-access articles still have readable PMC deposits. If that fails, search the exact title for a bioRxiv, medRxiv, or arXiv preprint, try the publisher PDF directly, or ask the user for the PDF.

Can I download supplementary files from PMC with a script?▼

No. PMC serves supplementary files through a JavaScript proof-of-work interstitial, so command-line requests return a small placeholder page instead of the file. Get supplements through an interactive browser or ask the user for them.

Why does Europe PMC fullTextXML return 404 for some papers?▼

The fullTextXML endpoint only covers the open-access subset and 404s for non-OA records. That 404 does not mean the paper is unavailable; fall through to the PMC HTML page, which works for many non-OA deposits.

Does this skill work on the UChicago RCC Midway cluster?▼

Yes. Midway login nodes are already on the University of Chicago network, so IP-authenticated subscription content resolves without a VPN. The real barrier is publisher bot detection like ScienceDirect captchas, which requires an interactive browser instead.

What are the limitations of reading figure captions instead of figures?▼

Captions do not show axes, distributions, image quality, lanes, or whether plotted data support the prose. When a claim lives in a panel, download the figure images with --figures or render PDF pages and inspect them visually.

Related Skills

Explore More Education & Research Skills→
K-Dense-AIK-Dense-AI
41.1k

paper-lookup

Search 11 academic literature APIs for papers, citations, and open-access full text with reproducible provenance.

Official
Advanced
s3644s3644
5

literature-review

Conducts systematic literature reviews with verified citations and PRISMA-compliant synthesis.

Community
Advanced
Jensen-YaoJensen-Yao
2

sci-figure

Extract figures and sub-figures from academic PDF papers at configurable DPI.

Community
Advanced
Yjj0333Yjj0333

nature-downloader

Downloads academic full text via publisher APIs, open-access routes, CNKI, and institution-authorized browser sessions.

Community
Advanced