nozomi-koborinai/gcp-cost-mcp-server

このページは AI により自動翻訳されています。技術的な詳細は原文をご確認ください。原文を見る

23 stars 2 forks 品質 79 トレンド 14MIT

💰 An MCP server that enables AI assistants to estimate Google Cloud costs, powered by Cloud Billing Catalog API and built with Genkit for Go

概要

: This is an unofficial project and is not affiliated with, endorsed by, or supported by Google or the Genkit team. An MCP (Model Context Protocol) server for estimating Google Cloud running costs. Instead of manually using the Google Cloud Pricing Calculator, you can get GCP cost estimates directly from AI assistants like Claude Desktop, Gemini CLI, or Cursor. Each tool is . AI assistants autonomously decide which tools to call and in what order based on context: show data flow — one tool's output feeds another's input. get_estimation_guide and estimate_cost are : they internally resolve their own dependencies, reducing round-trips.

README

以下は原文 README の抜粋です。原文を見る。

GCP Cost MCP Server

Note: This is an unofficial project and is not affiliated with, endorsed by, or supported by Google or the Genkit team.

An MCP (Model Context Protocol) server for estimating Google Cloud running costs.

Instead of manually using the Google Cloud Pricing Calculator, you can get GCP cost estimates directly from AI assistants like Claude Desktop, Gemini CLI, or Cursor.

Features

Available Tools

Tool	Description
`get_estimation_guide`	Start here! Dynamically generates estimation guides from SKU analysis for any GCP service
`classify_resource_cost`	Classifies Terraform resource billing behavior and estimates an authorized-count baseline
`list_services`	Lists all available Google Cloud services with their IDs
`list_skus`	Lists SKUs (billable items) for a specific service
`get_sku_price`	Gets pricing details for a specific SKU
`estimate_cost`	Calculates cost based on SKU and usage amount, with automatic free tier deduction

Tool Relationships

Each tool is independent and stateless. AI assistants autonomously decide which tools to call and in what order based on context:

graph TB
    guide["get_estimation_guide─────────────────IN: service_nameOUT: required params, pricingfactors, free tier info, tips─────────────────Internally resolvesservice &amp; SKU lookup"]

    classify["classify_resource_cost─────────────────IN: Terraform resource type,resource attributesOUT: billing trigger, SKU,warnings, monthly estimate"]

    services["list_services─────────────────IN: name filter (opt)OUT: service_id, display_name"]

    skus["list_skus─────────────────IN: service_id, region, keywordOUT: sku_id, display_name,categories, regions"]

    price["get_sku_price─────────────────IN: sku_id, currencyOUT: price/unit, pricing tiers"]

    cost["estimate_cost─────────────────IN: sku_id, usage_amountOUT: estimated cost withautomatic free tier deduction─────────────────Internally resolves pricing"]

    services -- "service_id" --> skus
    skus -- "sku_id" --> price
    skus -- "sku_id" --> cost

Solid arrows show data flow — one tool’s output feeds another’s input. get_estimation_guide and estimate_cost are self-contained: they internally resolve their own dependencies, reducing round-trips.

Use Case	How AI Uses the Tools
Quick estimate	`get_estimation_guide` → gather user requirements → `estimate_cost`
Multi-service	Multiple `get_estimation_guide` + `estimate_cost` calls in parallel
Explore pricing	`list_services` → `list_skus` → `get_sku_price`

GitHub で README 全文を見る

インストール

npx @modelcontextprotocol/inspector ./gcp-cost-mcp-server

MCP 初心者？mcpskill.org のガイドから始めましょう →

設定

{
  "mcpServers": {
    "gcp-cost": {
      "command": "/opt/homebrew/bin/gcp-cost-mcp-server"
    }
  }
}