leie

Is this healthcare provider barred from federal health programs? Screening against the HHS OIG List of Excluded Individuals/Entities — the check you make before paying, employing or contracting with a provider.

Part of Pipeworx — an MCP gateway connecting AI agents to 1573+ live data sources.

Tools

Tool	Answers
check_provider_exclusion	"Is this provider on the OIG exclusion list?" — candidates with the fields that tell same-named people apart.
provider_exclusion_coverage	How many exclusions are held, how current the copy is, and why NPI often cannot disambiguate.

Auth

None. No key, no account.

Data source

HHS Office of Inspector General — List of Excluded Individuals/Entities (LEIE), which OIG publishes as a CSV for loading into your own system:

• https://oig.hhs.gov/exclusions/exclusions-list/
• https://oig.hhs.gov/exclusions/downloadables/UPDATED.csv

US federal public domain. 83,782 exclusions as loaded — 80,355 individuals and 3,427 businesses.

This names real people — read this before using it

A false positive here accuses somebody of healthcare fraud. The data makes that easy to do by accident:

• 690 excluded people share the surname SMITH. 591 share JOHNSON.
• Four different excluded men are called some form of John Smith — born 1949 (California, internal medicine), 1951 (Colorado, counselor), 1960 (New York, plastic surgery) and 1970 (Pennsylvania, general practice).

So check_provider_exclusion never returns a boolean. It returns candidates with date_of_birth, state, city, specialty and npi, and a match_confidence describing what was actually matched:

match_confidence	Meaning
npi_match	The NPI matched. An NPI identifies one provider — this is an identification.
exact_name_only	The full name matched exactly. Not an identification; several excluded people can share a name.
strong_name_similarity	Very close but not identical. A lead to check.
weak_name_similarity	Loosely similar. Most results at this level are different people.

Pass npi whenever you have it — it is the difference between a lead and an identification. Only 8,835 of 83,782 exclusions carry one, though, so for most records date of birth is the practical discriminator.

Currently excluded, not historically

Everyone on this list is currently excluded. The file has a reinstatement column and it is empty on every row, because OIG removes a reinstated party rather than flagging them. The ingest therefore deletes rows absent from each new file — "no longer listed" is how a reinstatement reaches you, and there is no expired state to read.

What this does not cover

State Medicaid exclusion lists, SAM.gov debarment, and licensure actions that have not led to an OIG exclusion. A clear result here is not a complete background check, and check_provider_exclusion says so on every no-match response.

Refreshing

Monthly, when OIG republishes:

node scripts/ingest-leie.mjs

Re-runnable: rows upsert on a content hash, and anything absent from the new file is deleted so reinstatements take effect. It refuses to load a file with fewer than 50,000 rows rather than quietly emptying the table.

Quick Start

Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):

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

What this endpoint actually serves

tools/list at https://gateway.pipeworx.io/leie/mcp returns the tools in the table above plus the shared Pipeworx meta-tools — ask_pipeworx, discover_tools, search_within, remember/recall and the rest of the gateway-wide set. So the tool count you see is larger than this table: a single-pack endpoint currently lists roughly 30 shared tools alongside the pack's own. The connection's initialize response states its exact scope, and is the authoritative answer for a given day.

This is deliberate, not multiplexing by accident. The meta-tools are what let a scoped connection answer a question this pack does not cover — via ask_pipeworx, which routes across the whole catalog — without you adding a second MCP server. There is currently no way to mount a pack endpoint without them; if the extra schemas cost you more context than the routing is worth, connect to the full gateway once rather than to several pack endpoints.

Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:

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

Both URLs reach the same gateway and the same 1573+ data sources. The only difference is which pack's tools are listed directly; ask_pipeworx reaches all of them from either one.

Standalone (no gateway account)

This package also runs as a local stdio MCP server — no Pipeworx account, no gateway round-trip:

{
  "mcpServers": {
    "leie": {
      "command": "npx",
      "args": ["-y", "@pipeworx/mcp-leie"]
    }
  }
}

Or run it directly to confirm it starts:

npx -y @pipeworx/mcp-leie

It speaks MCP over stdin/stdout and answers initialize/tools/list/tools/call for only this pack's tools — none of the shared meta-tools the gateway connection above adds. Same source, same tools, no ask_pipeworx routing.

Using with ask_pipeworx

Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:

ask_pipeworx({ question: "your question about Leie data" })

The gateway picks the right tool and fills the arguments automatically.

More

• Docs and guides
• pipeworx.io

License

MIT