Google Scholar Scraper

Scrape Google Scholar at scale. One actor, six modes - search research papers, pull citation formats, fetch author profiles with h-index and i10-index, paginate an author's full publication list, view per-article citation history, and map an author's full co-author network. Built for literature reviews, bibliometrics, citation tracking, and academic AI agents.

What this actor returns

• Research paper search results: title, link, snippet, authors, publication info, cited-by counts, and version links.
• Citation strings for any paper in MLA, APA, Chicago, Harvard, and Vancouver formats, plus BibTeX / EndNote / RefMan / RefWorks export links.
• Full author profiles with name, affiliations, email domain, interests, photo, and the standard h-index / i10-index / total-citations table (overall and recent window).
• Year-by-year citation history graphs for both authors and individual papers.
• Author publication list, paginated up to 100 results per page.
• Per-article bibliographic detail including journal, volume, issue, pages, publisher, and abstract.
• Full co-author list with profile URLs, affiliations, email domains, and photos.

The six modes

Choose with the mode input parameter.

mode value	What it does	Required fields
search	Search Google Scholar for papers	q (or cites, or cluster)
cite	Get citation formats and BibTeX export for one paper	result_id
author_profile	Fetch an author's profile + citation metrics + graph	author_id
author_articles	Paginate the author's full publication list	author_id
author_citation	Per-article bibliographic record with citation history	author_id, citation_id
author_co_authors	Full list of an author's co-authors	author_id

Example: search

{
  "mode": "search",
  "q": "transformer neural network",
  "as_ylo": 2020,
  "as_yhi": 2024,
  "num": 10,
  "max_pages": 3
}

Example: cite

{
  "mode": "cite",
  "result_id": "K7uerNYAAAAJ:u5HHmVD_uO8C"
}

Example: author_profile

{
  "mode": "author_profile",
  "author_id": "LSsXyncAAAAJ"
}

Example: author_articles

{
  "mode": "author_articles",
  "author_id": "LSsXyncAAAAJ",
  "sort": "pubdate",
  "num": 20,
  "max_pages": 2
}

Example: author_citation

{
  "mode": "author_citation",
  "author_id": "LSsXyncAAAAJ",
  "citation_id": "u5HHmVD_uO8C"
}

Example: author_co_authors

{
  "mode": "author_co_authors",
  "author_id": "LSsXyncAAAAJ"
}

Input parameters

Parameter	Type	Modes	Description
mode	string	all (required)	Which operation to run.
q	string	search	Free-text search. Supports author: and source: operators.
cites	string	search	Find papers that cite this article ID.
cluster	string	search	Fetch all versions of a paper by cluster ID.
result_id	string	cite	Result ID of a paper to fetch citation formats for.
author_id	string	author_profile, author_articles, author_citation, author_co_authors	Google Scholar author identifier.
citation_id	string	author_citation	Per-article ID within an author's profile.
hl	enum	all	UI language (en, es, fr, de, ...).
lr	string	search	Restrict to languages, e.g. lang_en|lang_fr.
as_ylo	integer	search	Earliest publication year.
as_yhi	integer	search	Latest publication year.
scisbd	enum	search	0 relevance, 1 abstracts-only by date, 2 all by date.
as_sdt	enum	search	0 exclude patents, 7 include patents, 4 case law.
safe	enum	search	active / off.
filter	enum	search	1 enable similar-results filter (default), 0 disable.
as_vis	enum	search	0 include citations (default), 1 exclude citations.
as_rr	enum	search	1 review articles only, 0 all (default).
sort	enum	author_profile, author_articles	title / pubdate. Omit for default citation-count sort.
max_pages	integer	search (1-20 per page), author_articles (1-100 per page)	Max pages to fetch. 0 = no limit. Default 1.
num	integer	search, author_articles	Per-page size.

Example output (mode=search)

{
  "_mode": "search",
  "_query_index": 1,
  "search_parameters": { "mode": "search", "q": "transformer", "as_ylo": 2020 },
  "search_metadata_status": "Success",
  "search_timestamp": "2026-05-13T20:00:00Z",
  "position": 0,
  "result_id": "K7uerNYAAAAJ:u5HHmVD_uO8C",
  "paper_title": "Attention Is All You Need",
  "link": "https://arxiv.org/abs/1706.03762",
  "snippet": "...",
  "publication_info": {
    "summary": "A Vaswani, N Shazeer, N Parmar - Advances in NIPS, 2017",
    "authors": [
      { "name": "Ashish Vaswani", "author_id": "..." }
    ]
  },
  "inline_links": {
    "cited_by_total": 120000,
    "cited_by_link": "https://scholar.google.com/...",
    "versions_total": 95,
    "versions_cluster_id": "13755340029141322000"
  }
}

Example output (mode=author_profile)

{
  "_mode": "author_profile",
  "_query_index": 1,
  "author": {
    "name": "Geoffrey Hinton",
    "affiliations": "Emeritus Prof. Comp Sci, University of Toronto",
    "email": "Verified email at cs.toronto.edu",
    "interests": [
      { "title": "Machine Learning", "link": "..." }
    ]
  },
  "cited_by_summary": {
    "citations_all": 800000,
    "citations_recent": 500000,
    "h_index_all": 150,
    "i10_index_all": 380,
    "recent_since_year": 2020
  },
  "cited_by_graph": [
    { "year": 2018, "citations": 35000 },
    { "year": 2019, "citations": 45000 }
  ]
}

Pricing

Pay-per-event. No subscription.

• Setup: $0.02 per run (charged once).
• Query executed: $0.02 per upstream call. For paginated modes (search, author_articles), that is once per page. For single-shot modes (cite, author_profile, author_citation, author_co_authors), that is once per run.

Worked examples:

• mode=search with max_pages=5 -> $0.02 setup + 5 * $0.02 = $0.12.
• mode=author_profile -> $0.02 setup + 1 * $0.02 = $0.04.
• mode=author_articles with max_pages=3, num=100 (full author bibliography) -> $0.02 + 3 * $0.02 = $0.08.

Use cases

• Build a literature review: search for a topic, then loop through organic_results[].result_id to pull citation strings via mode=cite.
• Track citation growth: run mode=author_profile on a watchlist of researchers and store the cited_by_graph over time.
• Map a research community: take any author_id and run mode=author_co_authors to harvest the full collaborator network.
• Bibliometric analysis: page through an author's entire publication list with mode=author_articles and max_pages=0 for unlimited.
• AI agents and RAG pipelines: feed structured Google Scholar JSON straight into a knowledge graph or vector store.

How to get started

1. Open the actor in the Apify console and click Try for free.
2. Pick a mode and fill in the required fields shown above.
3. Click Run.
4. Results land in the default dataset. Download as JSON, CSV, or Excel from the Storage tab, or use the Apify API.

You can also call this actor from your code via the Apify SDK (Python, JavaScript, or curl) or as a tool in any MCP-aware AI agent.

FAQ

Q: Why is my run charged for setup even when there are no results? The $0.02 setup fee covers the run's instance provisioning. If you only want results, set tight inputs (max_pages=1, narrow query) so the setup is the only charge.

Q: How do I find a result_id or author_id? Run mode=search first. Each item in the output contains result_id (use it for mode=cite or as a cites / cluster value) and publication_info.authors[].author_id (use it for any author mode).

Q: What languages are supported? The hl enum exposes the most common 29 languages. The upstream API supports more; if you need one that is not in the list, open an issue and we will add it.

Q: Why doesn't pagination always reach max_pages? Google Scholar stops returning results when it runs out of matches. Pagination ends early when the upstream API returns fewer items than num or signals no next page.

Q: My run failed with an authentication error. The actor needs an API key configured by the publisher. If you see "Missing SERPAI_KEY", the deployment is misconfigured - please report it.

Links

• Source code on GitHub: https://github.com/johnvc/ApifyGoogleScholar
• More scrapers from this publisher: https://apify.com/johnvc

Last Updated: 2026.05.17