Superpowers/Catalog/Research paper search API

Research paper search API

Find scholarly papers by title, authors or keywords and get the DOI, authors (with ORCID when registered), year and publisher for each, from Crossref's metadata on more than 180 million works. Resolve a DOI to its record the same way. Free, from the API or your AI agent.

Create an account, get $5 Provider details

Free per callFailed calls are freeProvider available nowREST API and MCP

Real example

Call made on 10/08/2026 through Crossref, answered in 1658 ms.

Request

curl -X POST https://superpowers.tools/v1/run \
  -H 'Authorization: Bearer $SUPERPOWERS_KEY' \
  -H 'content-type: application/json' \
  -d '{"tool":"crossref.search-works","idempotency_key":"example-1","input":{"query":"Deep learning LeCun Bengio Hinton Nature 2015","limit":2}}'

Normalized response

{
  "results": [
    {
      "doi": "10.1038/nature14539",
      "title": "Deep learning",
      "url": "https://doi.org/10.1038/nature14539",
      "authors": [
        {
          "name": "Yann LeCun",
          "orcid": null
        },
        {
          "name": "Yoshua Bengio",
          "orcid": null
        },
        {
          "name": "Geoffrey Hinton",
          "orcid": null
        }
      ],
      "year": 2015,
      "publisher": "Springer Science and Business Media LLC"
    },
    {
      "doi": "10.1145/3448250",
      "title": "Deep learning for AI",
      "url": "https://doi.org/10.1145/3448250",
      "authors": [
        {
          "name": "Yoshua Bengio",
          "orcid": null
        },
        {
          "name": "Yann Lecun",
          "orcid": null
        },
        {
          "name": "Geoffrey Hinton",
          "orcid": null
        }
      ],
      "year": 2021,
      "publisher": "Association for Computing Machinery (ACM)"
    }
  ]
}

Providers

Same input and output for every provider. Success rate over 30 days shows after 5 real calls.

Provider	Price per call	Success	Status
Crossref	Free	—	Available

Returned fields

`results[].doi / url`	The paper's DOI and its doi.org link
`results[].title`	Registered title
`results[].authors`	Author names, with ORCID iD when the publisher registered one (up to 30)
`results[].year / publisher`	Publication year and publisher

Use it from your AI agent

Add the MCP server to Claude, Cursor or another client and ask in plain language.

MCP configuration

{
  "mcpServers": {
    "superpowers": {
      "url": "https://superpowers.tools/mcp",
      "headers": {
        "Authorization": "Bearer sp_live_…"
      }
    }
  }
}

Ask your agent

“Find the 2015 Nature review on deep learning by LeCun, Bengio and Hinton and give me its DOI and citation.”

Frequently asked questions

How is the search ranked?

By how well the title, authors and other bibliographic text match your query, not by citations or meaning. Include the title words plus an author or year for the best first result; a short generic phrase can surface newer papers with similar titles.

Can I look up a DOI directly?

Yes. crossref.lookup-doi takes a DOI such as 10.1038/nature14539 and returns the registered title, authors, year and publisher. Resolving 10.1038/nature14539 took about 1 second on 2026-10-08.

Does it return the full text or abstract?

No. It returns publisher-registered metadata. Follow the doi.org link for the paper itself, which may be paywalled.

How many results per call?

Up to 5 (limit, default 3). Both tools are free; you need an account for an API key, and the $5 welcome credit covers the paid tools.

Can my AI agent use it?

Yes. Add the Superpowers MCP server and your agent can find papers, check DOIs and build citations on its own, for example to verify that a reference it is about to quote exists.

Start with $5 of credit

This call is free; the credit covers the paid tools. No card needed to start.

Create an account