Skip to main content
Glama
namegender-mcp

namegender-mcp

NameGender MCP server

Glama quality score

Turns names, email addresses and usernames into a gender — with the evidence next to every answer: the probability, the sample size, the source and the name it matched.

Works in any client that speaks the Model Context Protocol: Claude Desktop, Claude Code, Cursor and others.

namegender-mcp MCP server – quality and maintenance score on Glama

Setup

Get an API key from the namegender.com dashboard. Sign-up needs no card and comes with free credits every day.

Claude Desktop / Claude Code configuration:

{
  "mcpServers": {
    "namegender": {
      "command": "npx",
      "args": ["-y", "namegender-mcp"],
      "env": { "NAMEGENDER_API_KEY": "ng_live_..." }
    }
  }
}

Claude Code from the command line:

claude mcp add namegender -e NAMEGENDER_API_KEY=ng_live_... -- npx -y namegender-mcp

Related MCP server: Email Social Media Checker

Tools

Tool

What it does

gender_from_name

Gender from a first or full name

gender_from_email

Extracts a name from the local part of an email address

gender_from_username

Extracts a name from a username or handle

gender_bulk

Up to 100 values in one request, with a match-rate summary

name_countries

Countries a name is recorded in — not a claim about origin

account_status

Remaining credits, daily free quota, data version

The gender tools take an optional two-letter country code. The same name can have a different gender from one country to the next; with a code, the answer is weighted by that country's data.

What an answer looks like

Ayşe: female · probability 99% · sample 12,345 · source ssa · country TR

An answer from a source without counts says so:

Kamon: male · probability 95% · no sample (unverified) · source wgnd

The distinction is deliberate. A 95% backed by counted people and a 95% backed by nothing are not the same thing and should not look the same.

Countries are not origin

name_countries keeps two lists apart. Counted birth registrations are published for only seven countries (US, UK, France, Canada, Spain, Ireland, Norway), so the ranked shares compare those seven only. Every other country where the name is recorded appears in a separate, unranked list. Reading the ranking as "where the name comes from" is wrong: Mehmet ranks first in France there, and Turkey is not in the ranking at all.

Errors

Errors do not break the session. The tool returns isError and says what to do — where to top up when credits run out, which variable to check when the key is invalid — and includes the request ID so support can trace it.

Environment variables

Variable

Default

NAMEGENDER_API_KEY

(required)

NAMEGENDER_BASE_URL

https://namegender.com/api/v1

Development

npm install
npm test

MIT licensed.

Releasing

npm version patch   # also updates server.json
git push --follow-tags

The v* tag runs .github/workflows/release.yml: tests, then npm (trusted publishing, no token), then the MCP registry (GitHub OIDC). Re-running the workflow skips whatever is already published.

Available Tools

6 tools
account_statusAccount and remaining creditsA

Returns remaining credits, the daily free quota and the data version. When credits run out the other tools fail with 402; call this to find out why.

Inputs: no named parameters.

gender_bulkMany names at onceA

Resolves up to 100 values in one request and returns a summary with the match rate. Use this for more than one name: doing the same work one call at a time is slower and wasteful.

Inputs: names (required), type, country.

gender_from_emailGender from an email addressA

Extracts a first name from the local part of an email address and predicts its gender. Returns "unknown" when no name can be extracted — role addresses such as info@ or admin@ included.

Inputs: email (required), country.

gender_from_nameGender from a nameA

Predicts gender from a person's name. The answer carries more than the gender: the probability, how many people the sample is based on and how the name was matched. Names it cannot be sure about return "unknown" instead of a guess.

Inputs: name (required), country.

gender_from_usernameGender from a usernameA

Extracts a first name from a username or social media handle and predicts its gender.

Inputs: username (required), country.

name_countriesCountries a name appears inA

Returns the countries where a name is recorded. WARNING: this is NOT a claim about origin or ethnicity. Counted birth registrations are published for only seven countries (US, UK, France, Canada, Spain, Ireland, Norway), so the ranking compares those countries only; countries that publish no counts, such as Turkey or Japan, appear in the attested list, not in the ranking.

Inputs: name (required), limit.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.2/5.0

    Scored across 6 tools

    Disambiguation5/5

    Each tool targets a distinct input source or function: name, email, username, bulk processing, country lookup, and account status. No overlap in purpose, and the bulk tool is clearly differentiated as a batch alternative.

    Naming Consistency4/5

    The gender_* tools share a consistent prefix and verb_from_source pattern (gender_from_name, gender_from_email, gender_from_username, gender_bulk). name_countries and account_status deviate from the verb_noun pattern but still use snake_case and are descriptive, so the naming is mostly consistent with minor exceptions.

    Tool Count5/5

    Six tools is well-scoped for a gender prediction service: three single-input methods, a bulk method, a supplementary lookup, and a status/meta tool. Each tool serves a clear purpose without redundancy.

    Completeness5/5

    The surface covers all core operations: single prediction for name/email/username, bulk resolution, country information, and account status. There are no obvious missing operations for the domain, and the bulk tool addresses efficiency needs.

    Related MCP Connectors

    Related MCP Servers