philpapers-mcp

An MCP server wired to PhilPapers / PhilArchive — the philosophy preprint archive (the field's closest analog to arXiv). It lets an MCP client search philosophy papers, read their metadata and abstracts, browse recent submissions, and download open-access PDFs.

No API key required. Everything runs against keyless public endpoints.

How it works

| Tool | What it does | Backend | | --- | --- | --- | | search_papers | Keyword search (title/abstract/full text), returns metadata + PhilArchive links + PDF URL | OpenAlex, filtered to the PhilPapers Foundation source | | research | One-shot scan: search and pull each hit's verbatim full abstract + subjects in a single call | OpenAlex + OAI-PMH GetRecord | | get_paper | Canonical metadata + full abstract for one record id | PhilArchive OAI-PMH GetRecord | | list_recent | Records added/updated in a date window | PhilArchive OAI-PMH ListRecords | | fetch_pdf | Download a record's open-access PDF to disk, return the path | philpapers.org/archive/<ID>.pdf | | get_fulltext | Download the open-access PDF and return its extracted full text | PDF + unpdf text extraction |

PhilArchive is the open-access archive built on the PhilPapers database, so a record id such as BROTNO-9 resolves on both philarchive.org and philpapers.org.

Why OpenAlex for search? PhilPapers' own JSON search API needs a (free) key and sits behind Cloudflare, and OAI-PMH is harvest-only (no keyword search). OpenAlex indexes ~80k PhilPapers works with full-text search, abstracts, and links straight back to the PhilArchive record and PDF.

Example

search_papers with { "query": "phenomenal consciousness higher-order", "open_access_only": true, "limit": 3 }:

Found 1,806 match(es) in PhilPapers/PhilArchive; showing 3 (open-access only).

1. What Is Wrong with the No-Report Paradigm and How to Fix It (2019)
   id: BLOWIW-2
   authors: Ned Block
   philarchive: https://philarchive.org/rec/BLOWIW-2
   pdf: https://philpapers.org/archive/BLOWIW-2.pdf
   doi: https://doi.org/10.1016/j.tics.2019.10.001
2. The HOROR theory of phenomenal consciousness (2014)
   id: BROTNO-9
   authors: Richard Brown
   philarchive: https://philarchive.org/rec/BROTNO-9
   pdf: https://philpapers.org/archive/BROTNO-9.pdf
   doi: https://doi.org/10.1007/s11098-014-0388-7
...

Then get_fulltext with { "id": "BROTNO-9" } returns the paper's extracted full text (e.g. "14 page(s), 38,302 chars"), or fetch_pdf saves the PDF locally for the client to read.

For a ready-to-read digest in one call, research with { "query": "HOROR higher-order consciousness", "limit": 2 } returns each hit with its verbatim full abstract and subjects — no per-paper get_paper follow-ups needed.

Setup

Once published to npm, no clone or build is needed — run it straight with npx:

npx -y philpapers-mcp

Or from source:

git clone https://github.com/sea9401/philpapers-mcp
cd philpapers-mcp
npm install   # the `prepare` hook builds dist/ automatically

Register with Claude Code

# via npx (no clone)
claude mcp add philpapers -- npx -y philpapers-mcp

# from a local build
claude mcp add philpapers -- node /absolute/path/to/philpapers-mcp/dist/index.js

# optional: identify yourself to OpenAlex's "polite pool" for better rate limits
claude mcp add philpapers -e [email protected] -- npx -y philpapers-mcp

Then /mcp inside Claude Code should list philpapers with its four tools.

Register with Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "philpapers": {
      "command": "node",
      "args": ["/home/sea9401/philpapers-mcp/dist/index.js"],
      "env": { "OPENALEX_MAILTO": "[email protected]" }
    }
  }
}

Environment variables

| Var | Default | Purpose | | --- | --- | --- | | OPENALEX_MAILTO | [email protected] | Your email — joins OpenAlex's polite pool (recommended). | | PHILPAPERS_DOWNLOAD_DIR | <tmp>/philpapers-mcp | Where fetch_pdf saves files. |

Smoke test

node test-client.mjs

Spawns the server over stdio and exercises all four tools.

Notes & limits

• search_papers reaches both preprints and published papers that have a PhilPapers/PhilArchive record. open_access_only: true keeps only those with a downloadable PhilArchive PDF.
• list_recent filters on the OAI datestamp (when the record was added/updated), which can differ from the paper's publication date shown in the listing. It returns only the first OAI page (no resumption-token paging).
• Not every record has an open-access PDF — many entries are metadata-only links to the published version. get_paper reports PDF availability; fetch_pdf fails clearly when there isn't one.
• Abstracts from search_papers are reconstructed from OpenAlex's inverted index; get_paper returns the archive's verbatim abstract.

Publishing (maintainers)

CI (.github/workflows/ci.yml) builds on Node 18/20/22 for every push and PR.

To publish a new version to npm:

1. Add a repo secret NPM_TOKEN (an npm Automation access token) under Settings → Secrets and variables → Actions.
2. Bump the version and tag: npm version patch && git push --follow-tags.
3. Cut a GitHub Release — .github/workflows/publish.yml runs npm publish (with provenance) automatically.

Or publish manually: npm login then npm publish --access public.

License

MIT