`research_chain`

Pack: research-chain · Connect: https://pipeworx.io/mcp (see Connect below for a single-pack URL)

No MCP client? Call it directly: GET https://gateway.pipeworx.io/v1/tools/research_chain for the schema, then POST the same URL with its arguments for the data.

Everything openly available around a paper, in ONE call: full text, license, datasets, code, and citations. Given a DOI, PMID, or arXiv id, walks paper (Crossref + OpenAlex) -> open copies with PER-COPY license and version (Unpaywall + PubMed Central — never collapses “free to read” into “permitted to reuse”) -> datasets that cite it (DataCite relatedIdentifiers) -> code/materials that implement it (Zenodo, Hugging Face Papers) -> papers that cite it (Semantic Scholar) -> retraction status (Crossref/Retraction Watch). Use for “what datasets/code are linked to ”, “is this paper retracted”, “find an open copy of and its license”, “who has cited ”. Each leg is independently fetched — a slow/dead upstream is listed in sources_failed rather than failing the whole call.

Parameters

Name	Type	Required	Description
`doi`	string	no	DOI of the paper, e.g. “10.1038/s41586-020-2649-2” (a doi.org URL is also accepted).
`pmid`	string	no	PubMed ID, e.g. “32015507”. Resolved to a DOI via NCBI’s ID Converter.
`arxiv_id`	string	no	arXiv id, e.g. “2312.00752”. Resolved to arXiv’s own DataCite DOI (10.48550/arXiv.) — best-effort for pre-2022 papers.
`max_datasets`	number	no	Max datasets to return (1-100, default 25). DataCite’s `total` count is always returned even when truncated.
`max_citing`	number	no	Max citing papers to return (1-25, default 10).

Example call

Arguments

{
  "doi": "10.1038/s41586-020-2649-2"
}

curl

curl -X POST https://gateway.pipeworx.io/research-chain/mcp \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"research_chain","arguments":{"doi":"10.1038/s41586-020-2649-2"}}}'

TypeScript (@pipeworx/sdk)

import { Pipeworx } from '@pipeworx/sdk';
const pipeworx = new Pipeworx();

const result = await pipeworx.call('research_chain', {
  "doi": "10.1038/s41586-020-2649-2"
});

More examples

{
  "doi": "10.1145/3680528.3687680",
  "max_datasets": 10,
  "max_citing": 5
}

Connect

Add this to your MCP client config — every tool in the catalog, including this one — or use one-click install buttons:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://pipeworx.io/mcp"
    }
  }
}

Connect to just the research-chain pack

{
  "mcpServers": {
    "research-chain": {
      "url": "https://gateway.pipeworx.io/research-chain/mcp"
    }
  }
}

See Getting Started for client-specific install steps.

research_chain