Back to Directory/Developer Tools

io.github.cyanheads/npi-providers-mcp-server

Search NPPES providers and resolve NUCC specialty codes via MCP over STDIO or Streamable HTTP.

Developer ToolsTypeScriptv0.4.0

@cyanheads/npi-providers-mcp-server

Search NPPES providers and resolve NUCC specialty codes via MCP over STDIO or Streamable HTTP.

3 Tools • 2 Resources

Version License Docker MCP SDK npm TypeScript Bun

Install in Claude Desktop Install in Cursor Install in VS Code

Framework


Overview

US healthcare provider directory over the NPPES NPI Registry, with plain-language specialty terms resolved offline against a bundled NUCC taxonomy. Search providers by name, organization, location, and specialty; decode NPIs into full provider records; and resolve or browse the NUCC taxonomy directly. Runs as a stdio process, a local Streamable HTTP server, or the public hosted endpoint above.

Tools

ToolDescription
npi_search_providersSearch the NPPES registry by name, organization, location, provider type, and specialty. Plain-language specialties resolve through the bundled NUCC taxonomy before searching.
npi_get_providerFetch the complete NPPES record for up to 10 NPIs — taxonomies, addresses, credentials, identifiers, endpoints, and status.
npi_lookup_taxonomyResolve, fetch, or browse the NUCC Healthcare Provider Taxonomy — fully offline.

Resources

ResourceDescription
npi://provider/{npi}A single provider's full decoded record by NPI — the resource twin of npi_get_provider.
npi://taxonomy/{code}A single NUCC taxonomy entry by code — the resource twin of npi_lookup_taxonomy mode get.

All resource data is also reachable via tools; the resources are convenience twins for resource-capable clients.

Capability reference

npi_search_providers tool

  • Search by name_search shortcut, explicit first_name / last_name, organization_name, city / state / postal_code, and provider_type (individual / organization); at least one criterion is required and the registry rejects state-only searches
  • Plain-language specialty resolves through the bundled NUCC taxonomy to the registry's exact description before searching, echoed back via resolvedTaxonomies / appliedTaxonomyDescription; taxonomy_description is an escape hatch for an already-known exact description (mutually exclusive with specialty)
  • Trailing-wildcard (*) name/organization matching requires at least 2 leading characters
  • limit 1–200 (default 10), skip 0–1000; the registry never reports a true match total, only the first 1200 matches are reachable, and the response discloses page-size-not-total via truncated / notice
  • Typed error reasons: no_search_criteria, conflicting_specialty, unresolved_specialty, invalid_search_field

npi_get_provider tool

  • Accepts a single NPI or up to 10; each is validated as exactly 10 digits before any API call
  • Returns every taxonomy (with primary flag, license number and state), all practice and mailing addresses, credential, sex, sole-proprietor flag, enumeration and last-updated dates, secondary identifiers, and FHIR/Direct endpoints
  • Three-way partition: found (resolved records), notFound (confirmed absence — deactivated or never enumerated), errored (upstream failure, distinct from absence — retry these)
  • Throws none_found only when every requested NPI is a confirmed absence; an upstream failure on any NPI surfaces as that underlying error instead

npi_lookup_taxonomy tool

  • Three modes: resolve (plain-language term → matching codes/descriptions), get (exact code → full entry), browse (walk grouping → classification → specialization, filterable by grouping and NPI section)
  • resolve / browse cap results at limit (≤50, default 20) and disclose truncated; page past the cap with skip (0–1000, raised by limit each call)
  • A resolved entry's specialization (or classification when specialization is absent) is the exact value npi_search_providers.taxonomy_description accepts
  • Typed error reasons: no_match, missing_argument

npi://provider/{npi} resource

  • Returns the same fully decoded record as npi_get_provider, for one NPI, as application/json
  • npi must be a well-formed 10-digit NPI; no_record when the registry has none (deactivated or never enumerated)

npi://taxonomy/{code} resource

  • Returns the same entry as npi_lookup_taxonomy mode get, as application/json; cached publicly for 24 hours
  • code must match ^\d{3}[A-Z0-9]{6}X$; no_match when no entry exists for the code

Features

Built on @cyanheads/mcp-ts-core: stdio and Streamable HTTP transports, pluggable auth (none / jwt / oauth), swappable storage (in-memory, filesystem, Supabase, Cloudflare KV/R2/D1), structured logging with optional OpenTelemetry tracing.

NPI/NPPES-specific:

  • Keyless — runs against the public CMS NPPES NPI Registry API (v2.1) with no API key or account
  • Bundled NUCC taxonomy — the 879-code Healthcare Provider Taxonomy (v25.0) ships in the image and loads into an in-memory index at startup, so specialty resolution and code lookups work fully offline with no second upstream
  • Specialty resolution turns a vague term ("heart doctor") into the precise taxonomy description the registry filters on, and echoes the match back for the agent to verify
  • Detects the registry's quirk of returning HTTP 200 with an Errors[] body on validation failure and maps it to typed, recoverable error reasons

Agent-friendly output:

  • Provenance on search — the resolved taxonomy and the exact taxonomy_description sent to the registry are echoed back, so agents can see what was actually searched and re-run with a different code
  • Honest pagination — the returned count is disclosed as the page size, never a fabricated grand total, with the 1200-match reachable ceiling surfaced when a broad query is capped
  • Graceful partial failure — npi_get_provider returns per-NPI found / notFound / errored rows instead of failing the whole batch

Getting started

Public Hosted Instance

A public instance is available at https://npi-providers.caseyjhand.com/mcp — no installation required. Point any MCP client at it via Streamable HTTP, with this client config:

{
  "mcpServers": {
    "npi-providers-mcp-server": {
      "type": "streamable-http",
      "url": "https://npi-providers.caseyjhand.com/mcp"
    }
  }
}

Self-Hosted / Local

Add the following to your MCP client configuration file. No API key is required — the upstream NPPES registry is keyless.

{
  "mcpServers": {
    "npi-providers-mcp-server": {
      "type": "stdio",
      "command": "bunx",
      "args": ["@cyanheads/npi-providers-mcp-server@latest"],
      "env": {
        "MCP_TRANSPORT_TYPE": "stdio",
        "MCP_LOG_LEVEL": "info"
      }
    }
  }
}

Or with npx (no Bun required):

{
  "mcpServers": {
    "npi-providers-mcp-server": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@cyanheads/npi-providers-mcp-server@latest"],
      "env": {
        "MCP_TRANSPORT_TYPE": "stdio",
        "MCP_LOG_LEVEL": "info"
      }
    }
  }
}

Or with Docker:

{
  "mcpServers": {
    "npi-providers-mcp-server": {
      "type": "stdio",
      "command": "docker",
      "args": ["run", "-i", "--rm", "-e", "MCP_TRANSPORT_TYPE=stdio", "ghcr.io/cyanheads/npi-providers-mcp-server:latest"]
    }
  }
}

For Streamable HTTP, set the transport and start the server:

MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
# Server listens at http://localhost:3010/mcp

Prerequisites

  • Bun v1.4.0 or higher (or Node.js v24+).
  • No API key — the NPPES NPI Registry API is public and keyless. The NUCC taxonomy is bundled, so there is no second data source to provision.

Installation

  1. Clone the repository:
git clone https://github.com/cyanheads/npi-providers-mcp-server.git
  1. Navigate into the directory:
cd npi-providers-mcp-server
  1. Install dependencies:
bun install
  1. Configure environment (optional):
cp .env.example .env
# edit .env only if you need to override the NPPES base URL or timeout

Configuration

No required variables — the server runs out of the box against the keyless NPPES registry. All variables below are optional overrides.

VariableDescriptionDefault
NPPES_API_BASE_URLNPPES NPI Registry API base URL. Override for a private mirror or testing.https://npiregistry.cms.hhs.gov/api
NPPES_TIMEOUT_MSPer-request HTTP timeout for NPPES calls, in milliseconds.15000
MCP_TRANSPORT_TYPETransport: stdio or http.stdio
MCP_HTTP_PORTPort for the HTTP server.3010
MCP_AUTH_MODEAuth mode: none, jwt, or oauth.none
MCP_SESSION_MODEHTTP session handling: stateless, stateful, or auto. This server pins stateless; the schema default auto resolves to stateful.stateless
MCP_LOG_LEVELLog level (RFC 5424).info
STORAGE_PROVIDER_TYPEStorage backend.in-memory
OTEL_ENABLEDEnable OpenTelemetry instrumentation (spans, metrics, completion logs).false

See .env.example for the full list of optional overrides.

Running the server

Local development

  • Build and run:

    # One-time build
    bun run rebuild
    
    # Run the built server
    bun run start:stdio
    # or
    bun run start:http
    
  • Run checks and tests:

    bun run devcheck   # Lint, format, typecheck, security
    bun run test       # Vitest test suite
    bun run lint:mcp   # Validate MCP definitions against spec
    

Docker

docker build -t npi-providers-mcp-server .
docker run --rm -e MCP_TRANSPORT_TYPE=http -p 3010:3010 npi-providers-mcp-server

The Dockerfile defaults to HTTP transport, stateless session mode, and logs to /var/log/npi-providers-mcp-server. OpenTelemetry peer dependencies are installed by default — build with --build-arg OTEL_ENABLED=false to omit them.

Project structure

DirectoryPurpose
src/index.tscreateApp() entry point — registers tools and resources, inits services.
src/configServer-specific environment variable parsing and validation with Zod.
src/mcp-server/toolsTool definitions (*.tool.ts).
src/mcp-server/resourcesResource definitions (*.resource.ts).
src/services/nppesNPPES NPI Registry API client — query building, Errors[]-on-200 detection, normalization.
src/services/taxonomyBundled NUCC taxonomy service — in-memory index for resolve / get / browse.
tests/Unit and integration tests mirroring src/.

Development guide

See CLAUDE.md/AGENTS.md for development guidelines and architectural rules. The short version:

  • Handlers throw, framework catches — no try/catch in tool logic
  • Use ctx.log for request-scoped logging, ctx.state for tenant-scoped storage
  • Register new tools and resources in the createApp() arrays
  • Wrap external API calls: validate raw → normalize to domain type → return output schema; never fabricate missing fields

Data attribution

Provider data from the CMS NPPES NPI Registry (public domain). The bundled specialty codes are the NUCC Health Care Provider Taxonomy, © American Medical Association on behalf of the National Uniform Claim Committee (NUCC), redistributed unmodified beyond formatting under the NUCC permission. See NOTICE.

Contributing

Issues are welcome. Run checks and tests before submitting:

bun run devcheck
bun run test

License

Apache-2.0 — see LICENSE for details.

Installation

Source-derived launch command. Check the maintainer’s required arguments and credentials before running:

bash
npx -y @cyanheads/npi-providers-mcp-server

Set up in your AI client

Merge this template into ~/Library/Application Support/Claude/claude_desktop_config.json. Keep existing servers. Add any arguments, credentials, and permissions required by the maintainer; this template has not been install-tested.

json
{
  "mcpServers": {
    "io-github-cyanheads-npi-providers-mcp-server": {
      "command": "npx",
      "args": [
        "-y",
        "@cyanheads/npi-providers-mcp-server"
      ]
    }
  }
}

Restart Claude Desktop completely for changes to take effect. Confirm the server appears connected in the client’s tool list, then try a read-only example from its documentation.

Claude Desktop setup reference

Package

@cyanheads/npi-providers-mcp-servernpm

Compatible MCP Clients

io.github.cyanheads/npi-providers-mcp-server works with any MCP-compatible client. Copy the config snippet from the Configuration section above and add it to the file shown for your client, then restart the application.

  • Claude Desktop~/Library/Application Support/Claude/claude_desktop_config.jsonRestart Claude Desktop completely for changes to take effect.
  • Cursor~/.cursor/mcp.jsonRestart Cursor for changes to take effect.
  • VS Code.vscode/mcp.jsonReload VS Code window for changes to take effect.
  • Windsurf~/.codeium/windsurf/mcp_config.jsonRestart Windsurf for changes to take effect.
  • Claude Code.mcp.jsonSave at the project root, then start Claude Code in that project and review the MCP server approval prompt. Keep real credentials out of shared files.

Learn More