Nypl logo

Nypl

Searches NYPL Digital Collections and retrieves full item details through two MCP tools.

Install Nypl

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "nypl": {
      "command": "npx",
      "args": [
        "-y",
        "@pipeworx/mcp-nypl"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

We couldn’t confirm it starts correctly

The install command started, but didn't respond the way we expected when we tried to talk to it.

npx -y @pipeworx/mcp-nypl

initialize succeeded but no response to tools/list.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 6d ago.

Overview

The Nypl MCP server connects AI clients to the NYPL Digital Collections for keyword searches and item lookups. Its `search` tool returns matching records with identifiers, titles, creators or sources, dates, and links; `item` retrieves full details using an NYPL item UUID. Use the hosted Pipeworx endpoint for remote access or run the package locally as a stdio MCP server.

Use cases

  • Search NYPL Digital Collections by subject or keyword
  • Retrieve complete metadata for a collection item
  • Build a search-then-fetch research workflow
  • Ask Pipeworx to route questions about NYPL data

Key features

  • Keyword search across NYPL Digital Collections
  • Full item lookup by NYPL UUID
  • Hosted MCP access through Pipeworx
  • Local stdio package via npm
  • Optional gateway-wide natural-language routing

Tools2

What Nypl lets your agent do. Expand a tool for its parameters.

Self-reported~66 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 3d ago

Recent health check succeeded.

4/4 checks healthy over the last 44d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 3d ago via OSV.dev · @pipeworx/mcp-nypl (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars1Stargazers on the source repository.
npm downloads423Package downloads in the last 30 days.
Last commit1mo agoMost recent push to the default branch.
Install checkInconclusiveStarted but did not complete the MCP handshake in our test.
Tools exposed2Callable tools this server registers over MCP.
Directory activity1 viewsConfig copies, upvotes, and views on AllMCPs.

How Nypl works

What the Nypl MCP server does

The Nypl MCP server exposes search and retrieval capabilities for the NYPL Digital Collections. It is suited to agents that need to locate collection records by keyword and then inspect a particular record in more detail.

The search tool accepts a keyword query and returns matching items. Results include an item identifier, title, creator or source information, date data, and links. The returned identifier can be passed to the item tool.

The item tool retrieves the full details for one NYPL Digital Collections item. It expects an NYPL item UUID, normally obtained from a previous search call. This makes the two tools useful as a simple search-then-fetch workflow.

How it works

The Nypl MCP server is available in two forms. The hosted option uses the Pipeworx gateway at https://gateway.pipeworx.io/nypl/mcp. MCP clients connect to that URL and can call the NYPL tools remotely.

The gateway endpoint also exposes shared Pipeworx meta-tools, including ask_pipeworx, discover_tools, search_within, and remember/recall. These additional tools are part of the gateway connection and are not NYPL-specific. The endpoint may list roughly 30 shared tools in addition to the pack tools; its initialize response and tools/list response provide the current scope.

A local package is also available through npm. Running @pipeworx/mcp-nypl starts a stdio MCP server that handles initialize, tools/list, and tools/call for the NYPL pack only. The local mode does not add the gateway's shared meta-tools or require a Pipeworx account.

Setup and configuration

For a hosted connection, add the following MCP server configuration to a compatible client:

config.json
{
  "mcpServers": {
    "nypl": {
      "url": "https://gateway.pipeworx.io/nypl/mcp"
    }
  }
}

For local stdio use, configure the client to run npm with the package name:

config.json
{
  "mcpServers": {
    "nypl": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-nypl"]
    }
  }
}

You can verify the local process directly with npx -y @pipeworx/mcp-nypl. The hosted gateway can also be called over HTTP without an MCP client. For example, POST https://gateway.pipeworx.io/v1/tools/nypl_search with a JSON body such as {"query":"Civil War photographs"} invokes a search.

Tools and capabilities

  • search: Finds NYPL Digital Collections items by keyword and returns identifying and descriptive fields.
  • item: Fetches full information for one item using its NYPL UUID.
  • Hosted ask_pipeworx: Lets the gateway route a natural-language question to an appropriate tool across the broader Pipeworx catalog.

Use the local server when the agent only needs NYPL functionality and a local stdio process. Use the hosted endpoint when gateway access or its shared routing tools is useful.

Limitations and notes

The item operation depends on an NYPL item UUID, so a search result is the normal starting point. The hosted pack endpoint does not list only the two NYPL tools; shared Pipeworx tools are included by design. The local package avoids those extra tools but provides no ask_pipeworx routing. The hosted and local forms use the same NYPL source and expose the same pack-specific tools.

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Nypl

Nypl is an MCP server that connects AI clients to the NYPL Digital Collections. Its search tool finds collection records by keyword, while item retrieves full details for an NYPL item UUID returned by search. It can run locally over stdio or through Pipeworx's hosted MCP gateway.

Run it with `npx -y @pipeworx/mcp-nypl` and add that command to your MCP client's server configuration. The local package communicates over stdin/stdout and does not require a Pipeworx account.

Yes, Nypl works with Cursor through the hosted MCP URL or the local stdio configuration. Add the server entry to Cursor's MCP settings, using either `https://gateway.pipeworx.io/nypl/mcp` or the npm command.

First call `search` with a keyword, then pass an NYPL item UUID from the results to `item`. The search response includes identifiers and links that help select the record to fetch.

Related MCP servers

Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewNypl AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/nypl?style=directory)](https://allmcps.com/mcp/nypl)
HTML Embed
<a href="https://allmcps.com/mcp/nypl"><img src="https://allmcps.com/api/badge/nypl?style=directory" alt="Nypl on AllMCPs" /></a>