Skip to main content
Glama
Sign In
Guia da Maquininha MCP Server

Guia da Maquininha MCP Server

by canalthailexa-lgtm
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Hybrid

MCP Server - Guia da Maquininha

MCP Server que expõe dados reais de taxas de maquininha de cartão no Brasil (2026). Permite que modelos de IA consultem e comparem taxas de diferentes marcas (Stone, Cielo, Mercado Pago, PagBank, SumUp, InfinitePay, Ton, Getnet) para diferentes perfis de negócio.

Instalação

npm install

Related MCP server: tokenomics

Uso

O servidor roda via stdio e segue o Model Context Protocol padrão.

Ferramentas disponíveis

  1. get_brands - Lista todas as marcas com taxas MDR

  2. get_tiers - Lista perfis de negócio (MEI, pequeno, médio, grande)

  3. calculate_cost - Calcula custo mensal real para marca + perfil

  4. compare_brands - Compara todas as marcas e retorna ranking de custo

Exemplo de chamada

{
  "name": "compare_brands",
  "arguments": {
    "tier": "mei",
    "antecipation": false
  }
}

Dados

As taxas são baseadas em pesquisa de mercado 2026 com:

  • MDR crédito/débito por marca

  • Taxa de antecipação

  • Perfil de ticket médio e volume por tier de negócio

Fonte

guiamaquininha.com.br — comparador de maquininhas de cartão no Brasil. Dados originais de taxas MDR, simuladores de custo mensal e guias práticos para pequenos negócios escolherem a adquirente certa.

Registro MCP

Este servidor está pronto para submissão em registry.modelcontextprotocol.io

Registry MCP (publicado)

Available Tools

4 tools
calculate_costC

Calcula o custo mensal real de uma maquininha para um perfil específico

Inputs: tier (required), brand (required), antecipation.

compare_brandsA

Compara todas as marcas de maquininha para um perfil de negócio e retorna o ranking de custo

Inputs: tier (required), antecipation.

get_brandsA

Lista todas as marcas de maquininha disponíveis com suas taxas MDR

Inputs: no named parameters.

get_tiersB

Lista os perfis de negócio (MEI, pequeno, médio, grande) com ticket médio e volume

Inputs: no named parameters.

Tool Schema Changelog

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

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.5/5.0

    Scored across 4 tools

    Disambiguation4/5

    get_brands and get_tiers clearly target distinct resources (brands vs. business profiles), and calculate_cost vs. compare_brands differ in scope (single brand cost vs. cross-brand ranking). The two cost tools are conceptually related since ranking is an aggregation of per-brand costs, but the descriptions make the boundary workable.

    Naming Consistency5/5

    All four tools follow a clean verb_noun snake_case pattern: get_brands, get_tiers, calculate_cost, compare_brands. The verbs vary appropriately with the action while the structure stays fully predictable.

    Tool Count5/5

    Four tools is well-scoped for a focused card-machine cost comparison domain: two reference listers, one calculator, and one comparator. Each tool earns its place with no redundancy.

    Completeness4/5

    The surface covers the core workflow: list brands, list profiles, calculate cost for a profile, and rank brands. Minor gaps exist (no single-brand detail lookup or fee-type breakdown beyond MDR), but agents can still answer the domain's main questions without dead ends.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers