📚 OpenCitations Scraper: Citation Graph & DOI References

Overview

OpenCitations Scraper maps the citation network of any scientific paper. Give it one or more DOIs and get back every paper that cites the work (incoming citations), every paper the work cites (outgoing references), or both directions in a single run. Each record includes the citing DOI, the cited DOI, the citation date, and self-citation flags at both journal and author level. Powered by the OpenCitations INDEX (1.6 billion open citation links) with no API key required.

Until 2018, comprehensive citation data was locked behind Clarivate's Web of Science and Elsevier's Scopus at thousands of dollars per institutional seat. The OpenCitations INDEX changed that by assembling an open citation graph from Crossref reference deposits and other open sources, updated monthly. This actor wraps the OpenCitations INDEX API v2 into a clean, batch-capable scraper with automatic retry and rate-limit handling.

✅ No API key | ✅ 1.6B open citation links | ✅ Zero charge on empty runs | ✅ First 10 results free | ✅ MCP-ready for AI agents

Features

Look up citations, references, or both directions for any DOI in a single run. Batch input: pass an array of DOIs and the actor processes each with polite rate limiting. Self-citation detection: journal-level and author-level self-citation flags on every record. Citation dates at month or year precision, so you can plot citation velocity over time. Backward and forward reference chaining, the two classic snowballing methods for systematic reviews.

How it works

The actor calls the OpenCitations INDEX API v2 (api.opencitations.net/index/v2) for each DOI you supply. For direction citations, it retrieves every paper that cites your DOI (incoming). For references, it retrieves every paper your DOI cites (outgoing). For both, it runs both queries and merges the output.

Requests are paced with a 400ms delay between calls to respect the OpenCitations rate limit. DOIs that return no matches or a 404 log a debug message and continue without failing the run. A final summary record reports total citation counts and the run timestamp.

Reliability posture: blocked, empty, or failed runs are never charged. You only pay for a citation link that was actually delivered to the dataset.

🧾 Input configuration

{
  "dois": ["10.1038/nature12373", "10.1126/science.1231143"],
  "direction": "citations",
  "maxPerDoi": 200
}

📤 Output format

{
  "source_doi": "10.1038/nature12373",
  "direction": "references",
  "citing_doi": "omid:br/06120344846 doi:10.1038/nature12373 openalex:W2159974629 pmid:23903748",
  "cited_doi": "omid:br/062102024238 doi:10.1103/physrevx.2.031001 openalex:W3101024234",
  "citation_date": "2013-07-31",
  "journal_sc": "no",
  "author_sc": "no",
  "scraped_at": "2026-07-15T04:18:57.149Z"
}

Every citation link contains these fields:

Field	Description
🧬 source_doi	The input DOI this citation link belongs to
➡️ direction	citations, references, or both
📥 citing_doi	Identifier of the citing paper. OpenCitations returns a composite string (OMID, DOI, OpenAlex ID, PMID where available), not a bare DOI
📤 cited_doi	Identifier of the paper being cited, same composite format
📅 citation_date	Citation date at month or year precision
📰 journal_sc	yes if the citing and cited papers share a journal
👤 author_sc	yes if there is author overlap between the papers
🕒 scraped_at	ISO timestamp of capture

💼 Common use cases

Academic impact tracking Retrieve every paper that has cited your work or a seminal paper in your field without paying for Web of Science. Plot citation velocity over months and years using citation_date.

Systematic reviews Run backward reference chaining (what does this paper cite) and forward citation chaining (who has cited this paper since). Feed a seed set of DOIs and expand it into a snowball corpus in one run.

Scientometrics and bibliometrics Compute h-indexes, co-citation clusters, and citation velocity for funding bodies, universities, and publishers. Filter out self-citations using journal_sc and author_sc for cleaner impact metrics.

AI knowledge graphs Build structured citation edges between DOI nodes for graph neural networks or recommendation systems. Combine with OpenAlex for enriched node metadata (title, authors, venue, concepts).

🚀 Getting started

1. Open the actor and paste a list of DOIs in the DOIs field (e.g. ["10.1038/nature12373"]).
2. Pick a Citation direction: citations, references, or both.
3. Set Max results per DOI (1 to 2,000) to cap the fan-out for highly cited papers.
4. Click Save & Start.
5. Download as JSON, CSV, or Excel, or pull via API or MCP.

FAQ

Do I need an API key for OpenCitations? No. The OpenCitations INDEX API is fully open and requires no authentication. This actor works out of the box.

How complete is the citation coverage? OpenCitations INDEX is built from Crossref reference deposits and other open sources. Coverage is excellent for journals that participate in Crossref's reference deposit program (most major publishers since ~2010) and incomplete for older literature and some regional publishers.

What if a DOI has zero results? OpenCitations returns a 404 for DOIs not in its index. The actor logs a debug message and moves on to the next DOI without failing. Empty runs are never charged.

Can I look up multiple DOIs in one run? Yes. Provide an array of DOIs in the dois input. The actor processes each sequentially with a 400ms delay. For very large batches (100+ DOIs), consider splitting across multiple runs.

How am I charged? Pay per citation link at $0.003 per link. The first 10 results are free for life on every Apify account. Runs that return zero citations are never charged.

Use in Claude, ChatGPT & any MCP agent

https://mcp.apify.com/?tools=themineworks/opencitations-citation-graph

Or call it programmatically with the Apify client:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });

const run = await client.actor('themineworks/opencitations-citation-graph').call({
  dois: ['10.1038/nature12373'],
  direction: 'citations',
  maxPerDoi: 200,
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

🛠️ Complete your scholarly research pipeline

Citations are one edge type. Add the nodes and cross-references:

• OpenAlex Scraper: 250M+ papers with authors, venues, abstracts, and concepts.
• Crossref Scraper: DOI metadata for any published work from Crossref.
• ClinicalTrials.gov Scraper: cross-check biomedical citations with the underlying trial registry.

Typical flow: search OpenAlex for a topic, feed the top-cited DOIs into this actor for backward and forward chaining, then enrich the resulting node set with Crossref metadata.

Found a bug or have a feature request? Open an issue on the actor's Apify Console page or reach out through the Apify profile.