Wardn Hub
MCP ServersSkillsCategoriesAPI docsSubmit server
Submit server
Wardn HubTrusted MCP server directory.

Registry

  • MCP Servers
  • Skills
  • Categories

Resources

  • API docs
  • Score method

Contribute

  • Submit server
  • Advertise
© 2026 Wardn Hub
Wardn Hub
MCP ServersSkillsCategoriesAPI docsSubmit server
Submit server
skills/brycewang-stanford/Auto-Empirical-Research-Skills/43-wentorai-research-plugins-skills-research-funding-figshare-api

43-wentorai-research-plugins-skills-research-funding-figshare-api

1
brycewang-stanford/Auto-Empirical-Research-Skills·Research·Audit pending·Snapshot 9d1a731c9dd2

Summary

This source did not publish a separate summary. Review SKILL.md before using the skill.

SKILL.md

Figshare API Guide

Overview

Figshare is a cloud-based research data management platform that allows researchers to store, share, and discover research outputs including datasets, figures, media, papers, posters, and fileset collections. Every item uploaded to Figshare receives a citable DOI and is stored in a FAIR-compliant manner, making research outputs findable, accessible, interoperable, and reusable.

The Figshare API provides comprehensive programmatic access to the repository, enabling researchers and institutions to automate data publishing, integrate with research workflows, and build custom discovery interfaces. The platform supports versioning, embargo periods, and flexible access controls for both public and private research data.

Researchers, data managers, institutional repository administrators, and research infrastructure developers use the Figshare API to automate deposit workflows, harvest metadata for institutional dashboards, build data discovery tools, and integrate research data management into existing laboratory information management systems. Figshare serves over 150 institutions worldwide and hosts millions of research outputs.

Authentication

Authentication via personal access token is required for write operations and accessing private content. Read access to public content is available without authentication but has lower rate limits.

  1. Log in to Figshare at https://figshare.com/
  2. Navigate to Applications in your account settings
  3. Create a Personal Token with desired permissions
  4. Include the token in the Authorization header
curl -H "Authorization: token YOUR_FIGSHARE_TOKEN" "https://api.figshare.com/v2/account/articles"

Public endpoints can be accessed without a token for browsing published content.

Core Endpoints

articles: Search and Retrieve Articles

Search the public Figshare repository for published articles (datasets, figures, papers, media, and other item types).

  • URL: GET https://api.figshare.com/v2/articles
  • Parameters:
  • ParameterTypeRequiredDescription
    search_forstringNoFree-text search query
    item_typeintNoItem type filter (1=figure, 2=media, 3=dataset, etc.)
    published_sincestringNoFilter by date (YYYY-MM-DD format)
    orderstringNoSort: published_date, modified_date, views
    order_directionstringNoasc or desc
    pageintNoPage number (default 1)
    page_sizeintNoResults per page (default 10, max 1000)
    • Example:
    curl -X POST "https://api.figshare.com/v2/articles/search" \
      -H "Content-Type: application/json" \
      -d '{"search_for": "genomics CRISPR", "item_type": 3, "page_size": 5}'
    
    • Response: Returns array of article objects with id, title, doi, url, published_date, description, defined_type_name, categories, tags, authors, files (with download URLs), and citation.

    datasets: Manage Research Datasets

    Retrieve and manage dataset-specific content in Figshare. Datasets are a specialized article type with additional support for large file collections.

    • URL: GET https://api.figshare.com/v2/articles/{article_id}
    • Parameters:
    ParameterTypeRequiredDescription
    article_idintYesThe Figshare article/dataset ID
    • Example:
    # Get a specific dataset by ID
    curl "https://api.figshare.com/v2/articles/12345678"
    
    # List files in a dataset
    curl "https://api.figshare.com/v2/articles/12345678/files"
    
    • Response: Returns complete article object with id, title, doi, description, authors, categories, tags, files (array with name, size, download_url, computed_md5), license, version, is_embargoed, and custom_fields.

    Rate Limits

    Rate limits vary based on authentication status and endpoint. Authenticated requests generally allow up to 100 requests per minute. Unauthenticated requests are limited to approximately 10 requests per minute. The API returns HTTP 429 with a Retry-After header when limits are exceeded. For bulk data harvesting, Figshare provides OAI-PMH endpoints at https://api.figshare.com/v2/oai which are more suitable for large-scale metadata collection.

    Common Patterns

    Search for Datasets in a Research Area

    Find publicly available datasets matching specific research topics:

    import requests
    
    payload = {
        "search_for": "single cell RNA-seq",
        "item_type": 3,  # datasets only
        "page_size": 20,
        "order": "published_date",
        "order_direction": "desc"
    }
    
    resp = requests.post("https://api.figshare.com/v2/articles/search", json=payload)
    results = resp.json()
    
    for item in results:
        print(f"{item['title']}")
        print(f"  DOI: {item['doi']}")
        print(f"  Published: {item['published_date']}")
        print()
    

    Upload a Dataset Programmatically

    Automate data deposit for reproducible research workflows:

    import requests
    
    TOKEN = os.environ["FIGSHARE_API_TOKEN"]
    headers = {"Authorization": f"token {TOKEN}", "Content-Type": "application/json"}
    
    # Step 1: Create a new article
    article_data = {
        "title": "Supplementary Data for Analysis of Gene Expression",
        "defined_type": "dataset",
        "description": "RNA-seq counts and metadata for the analysis.",
        "tags": ["RNA-seq", "gene expression"],
        "categories": [69]  # Genetics category
    }
    resp = requests.post("https://api.figshare.com/v2/account/articles",
                         headers=headers, json=article_data)
    article_url = resp.json()["location"]
    
    # Step 2: Upload file
    article = requests.get(article_url, headers=headers).json()
    print(f"Created article ID: {article['id']}, DOI will be assigned on publish")
    

    Harvest Institutional Outputs via OAI-PMH

    Collect metadata from all Figshare items in an institution's repository:

    curl "https://api.figshare.com/v2/oai?verb=ListRecords&metadataPrefix=oai_dc&set=institution_123"
    

    References

    • Official API documentation: https://docs.figshare.com/
    • Figshare homepage: https://figshare.com/
    • API endpoint reference: https://docs.figshare.com/#figshare_documentation_api_description
    • OAI-PMH endpoint: https://docs.figshare.com/#oai_pmh

    Related skills

    r-reproducibility-guideAnswering Research QuestionsBuilding Paper Screening RubricsChina-CF-StudyCleaning Up Research Sessions