OpenCitations — Citation Graph (1.6 Billion Citation Links)

Map the citation network of any scientific paper from Apify. Given one or more DOIs, retrieve every paper that cites that work (incoming citations) and/or every paper it cites (outgoing references), with citation dates and DOIs for the full network. Powered by the OpenCitations INDEX — 1.6 billion open citation links — with no API key required.

Why This Actor?

Citation graphs are fundamental infrastructure for science and scholarship. Knowing who has cited a paper — and who those citers themselves cited — reveals the intellectual lineage of ideas, tracks the commercial adoption of academic discoveries, and exposes research communities invisible in keyword searches alone.

Until 2018, comprehensive citation data was locked behind Clarivate's Web of Science and Elsevier's Scopus, costing thousands of dollars per institutional license. The OpenCitations INDEX changed this by assembling an open, freely accessible citation graph from publisher deposits to Crossref and other open sources. It now covers 1.6 billion citation relationships and is updated monthly.

This actor wraps the OpenCitations INDEX API v2 (api.opencitations.net/index/v2) into a clean, batch-capable scraper with automatic retry and delay handling.

Target buyers and use cases:

• Academic researchers tracking impact who need every paper that has cited their work or a seminal paper in their field, without paying for Web of Science access
• Systematic review teams conducting backward reference chaining (what does this paper cite?) and forward citation chaining (who has cited this paper since?) — the two classic snowballing methods for systematic literature review
• Scientometrics analysts computing h-indexes, citation velocity curves, and co-citation clusters for funding bodies, universities, and publishers
• AI companies building knowledge graphs who need structured citation edges between DOI nodes to construct semantic research networks, train graph neural networks, or power recommendation systems
• Technology transfer offices tracking which commercial patent applications have cited academic papers from their institution

Natural Pairing with Other Actors

This actor works best as part of a two-step pipeline:

1. Start with a literature search actor (PubMed, arXiv, or a Crossref actor) to get a set of DOIs matching your research topic
2. Feed those DOIs into this actor to expand the network — retrieving all papers that cite each result (forward chaining) or all papers each result cites (backward chaining)

This two-step pattern replicates the snowballing methodology used in systematic reviews and can surface hundreds of relevant papers invisible to keyword search alone.

Citation Direction

Direction	What it returns
citations	Papers that cite your DOI — who has built on this work
references	Papers that this DOI cites — the intellectual sources this work draws on
both	Both directions in a single run

Inputs

Field	Type	Description	Default
dois	array	List of DOIs to look up (e.g. ["10.1038/nature12373"])	["10.1038/nature12373"]
direction	select	citations, references, or both	citations
maxPerDoi	integer	Maximum citation records per DOI (1–2,000)	200

Output Format

Each citation link is stored as one item in the Apify dataset:

{
  "source_doi": "10.1038/nature12373",
  "direction": "citations",
  "citing_doi": "10.1016/j.cell.2014.09.045",
  "cited_doi": "10.1038/nature12373",
  "citation_date": "2014-10",
  "journal_sc": "no",
  "author_sc": "no",
  "scraped_at": "2024-11-15T09:22:11.000Z"
}

Field notes:

• citing_doi — the DOI of the paper doing the citing
• cited_doi — the DOI of the paper being cited (your input DOI when direction is citations)
• citation_date — when the citing paper was published, at month or year precision
• journal_sc — whether the citation is self-citation at journal level (yes/no)
• author_sc — whether the citation is author self-citation (yes/no)

A summary record is appended at the end with total citation count and run timestamp.

Pricing

First 25 results are free on every Apify account — no charge until you exceed the free tier.

After the free tier: $3 per 1,000 citation links (Pay-Per-Event billing). A 1,000-link run costs $3.00. You are charged only for records actually delivered. Highly cited papers (Nature, Science, Cell) may return thousands of citations per DOI — use maxPerDoi to cap the output if you need a sample rather than the full network.

Frequently Asked Questions

Q: 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.

Q: 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. Web of Science remains more comprehensive for pre-2000 literature.

Q: What if a DOI has zero results? OpenCitations returns a 404 for DOIs not in its index. The actor handles this gracefully — it logs a debug message and moves on to the next DOI without failing the run.

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

Q: How do I find the DOI for a paper? DOIs are typically shown on the paper's journal page, in PubMed (doi field in output), or on doi.org. If you have titles or PMIDs from a PubMed search, run that actor first and use the doi field from those results as input here.

Q: What is the difference between journal_sc and author_sc? These are self-citation flags from OpenCitations. journal_sc is yes when the citing and cited papers are from the same journal. author_sc is yes when there is author overlap between the citing and cited papers. Self-citation detection is useful for impact analysis and bibliometric research.

Q: How often is the OpenCitations INDEX updated? OpenCitations updates its INDEX monthly from new Crossref reference deposits. Citation links for papers published in the last 1–2 months may not yet appear in the index.

Use in Claude, ChatGPT & any MCP agent

This actor is also a Model Context Protocol (MCP) server tool — call it directly from Claude, ChatGPT, Cursor, Windsurf, or any MCP-compatible AI agent. The agent only pays for results delivered (same pay-per-result model).

• Per-actor MCP endpoint: https://mcp.apify.com/?tools=themineworks/opencitations-citation-graph
• Full Mine Works MCP server (all tools): https://the-mine-works-mcp.hatchable.site/api/mcp

// Call this actor as a tool via apify-client (Node)
import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });
const run = await client.actor('themineworks/opencitations-citation-graph').call({ /* input from the table above */ });
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);