skills.rest
ClawBio
ClawBioOFFICIAL
Explore all skills by this author →
WebsiteGitHub

article-data-fetcher

Discovers and downloads genomics data files deposited in public repositories for a given article DOI or PMID.

1.1k|257|Updated Feb 25, 2026
Search Tags:#genomics#reproducibility#pubmed#data download#DOI#zenodo#gEO
One-click install
npx skills add https://github.com/ClawBio/ClawBio --skill article-data-fetcher
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: article-data-fetcher
Source: https://github.com/ClawBio/ClawBio/tree/main/skills/article-data-fetcher
Command: npx skills add https://github.com/ClawBio/ClawBio --skill article-data-fetcher

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires requests, biopython, tqdm, pydantic.

What problem does it solve?

Finding and downloading the raw genomics data deposited with a published paper requires manually hunting through GEO, ENA, Zenodo, and other repositories for accession numbers, then downloading files one by one — a process that can take 30–60 minutes per paper.

Core Features & Use Cases

  • Article Resolution: Resolves a DOI, PMID, PMCID, or repository URL into article metadata via Crossref, NCBI Entrez, and PMC, then scans for repository accessions (GSE, PRJNA, E-MTAB, Zenodo, Figshare, Dryad, OSF).
  • Interactive File Selection: Lists every available file with size and inferred data type, then downloads only the file types the user confirms, with MD5/SHA256 checksum validation and progress bars.
  • Auditable Outputs: Writes a machine-readable manifest.json (source URL, size, checksum per file) and a human-readable report.md summarizing what was found and downloaded.
  • Use Case: A researcher reads a single-cell paper and needs its H5AD matrix. They provide the DOI, confirm "h5ad" from the listed files, and the file lands locally with a verified checksum and full download manifest.

Quick Start

Ask the agent to download the VCF and FASTA files deposited with DOI 10.1038/s41586-021-03819-2 into a local downloads folder.

Frequently Asked Questions about article-data-fetcher

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

FAQPage Schema
How do I download genomics data files from a published paper?▼

Provide the paper's DOI, PubMed ID, or PMCID and the skill resolves the article, discovers linked repository accessions (GEO, ENA, Zenodo, and others), lists all available files, and downloads only the file types you confirm. Every download is checksum-validated and recorded in manifest.json.

What repositories can I download research data from using a DOI?▼

Supported repositories include NCBI GEO, SRA/ENA, ArrayExpress, Zenodo, Figshare, Dryad, and OSF. If no accession is found in the article text, fallback searches query DataCite, Zenodo, GEO, ENA, and Dryad by DOI, PMID, or title.

Can I download only specific file types like VCF or H5AD from a GEO accession?▼

Yes. After the file listing is shown, you can select files by number, range, extension (e.g. vcf,fasta,h5ad), or choose all. Nothing is downloaded until you confirm, and files larger than 10 GB in total trigger an extra size warning.

Does the skill verify downloaded file integrity?▼

Yes. When repositories provide MD5 or SHA256 checksums (GEO, ENA, and Zenodo do), each downloaded file is validated. If a checksum fails, the file is deleted and the failure is reported rather than passing corrupt data to the user.

What happens if a paper has no deposited data accession?▼

The skill reports clearly that no accession was found and suggests checking the paper's Data Availability Statement manually. It never invents accession numbers or guesses file listings.

What are the limitations of downloading supplementary data from publishers?▼

Some publishers host supplementary files behind paywalls even for open-access articles, returning HTTP 401/403. The skill detects these responses and informs the user instead of saving an HTML error page as if it were a data file.

Related Skills

Explore More Education & Research Skills→
cookjohncookjohn
19

pm-paper-detail

Extract structured metadata from PubMed articles by PMID.

Community
Intermediate
mims-harvardmims-harvard
1.7k

tooluniverse-single-cell

Analyze single-cell RNA-seq data with scanpy from QC gating through clustering and annotation.

Official
Advanced
omicverseomicverse
13

omicverse-micro-metabol-paired

Detect microbe-metabolite associations across matched samples using Spearman correlation, CCA, and MMvec workflows.

Official
Advanced
stahurastahura
19

data-api

Route Domo dataset reads through the dataset-query module.

Community
Intermediate