aws-cost-explorer

aws-cost-explorer

Enables AI assistants to query and analyze AWS costs and usage via the Cost Explorer API.

Category
访问服务器

README

AWS Cost Explorer MCP Server

A Python-based Model Context Protocol (MCP) server that provides tools for accessing AWS Cost Explorer data. This server enables AI assistants like Claude to query and analyze your AWS costs and usage.

Features

  • Cost and Usage Analysis: Get detailed cost breakdowns by time period
  • Dimension Values: Discover available services, regions, accounts, etc.
  • Flexible Filtering: Filter costs by various dimensions
  • Multiple Metrics: Support for different cost metrics (UnblendedCost, AmortizedCost, etc.)
  • Secure Authentication: Uses AWS profiles or environment variables

Prerequisites

  • Python 3.8 or higher
  • AWS account with Cost Explorer access
  • AWS credentials configured (via AWS CLI, environment variables, or IAM roles)

Installation

  1. Clone or download this repository

  2. Install dependencies:

    pip install -r requirements.txt
    
  3. Configure AWS credentials (choose one method):

    Option A: AWS CLI Profile

    aws configure --profile your-profile-name
    

    Option B: Environment Variables

    export AWS_ACCESS_KEY_ID=your_access_key
    export AWS_SECRET_ACCESS_KEY=your_secret_key
    export AWS_DEFAULT_REGION=us-east-1
    

    Option C: Use .env file

    cp .env.example .env
    # Edit .env with your credentials
    

Usage

Running the Server

python aws_mcp/main.py

The server will start and listen for MCP connections on stdio.

Connecting with Claude Desktop

Add the following to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "aws-cost-explorer": {
			"command": "uv",
			"args": [
				"--directory",
				"/absolute/path/to/aws_mcp",
				"run",
				"aws-cost-explorer",
				"--profile",
				"dev"
			],
    }
  }
}

Available Tools

1. get_cost_and_usage

Get AWS costs for a specified time period with optional grouping and filtering.

Parameters:

  • start (optional): Start date (YYYY-MM-DD, inclusive)
  • end (optional): End date (YYYY-MM-DD, exclusive)
  • granularity: Time granularity - "DAILY" or "MONTHLY" (default: "MONTHLY")
  • group_by (optional): List of dimensions to group by (max 2)
  • metrics (optional): List of cost metrics (default: ["UnblendedCost"])
  • filter_config (optional): Filter by dimension and values
  • max_results: Maximum results (1-100, default: 100)
  • profile (optional): AWS profile name

Example Usage:

"What were my AWS costs last month by service?"
"Show me daily costs for the last 30 days"
"Get costs by region for this year"

2. get_dimension_values

Get available values for a specific Cost Explorer dimension (services, regions, etc.).

Parameters:

  • dimension: The dimension name (SERVICE, REGION, LINKED_ACCOUNT, etc.)
  • time_period_start (optional): Start date for search
  • time_period_end (optional): End date for search
  • search_string (optional): Filter dimension values
  • max_results: Maximum results (1-1000, default: 50)
  • profile (optional): AWS profile name

Example Usage:

"What AWS services am I using?"
"Show me all available regions in my account"
"List my linked accounts"

Supported Dimensions

  • SERVICE
  • LINKED_ACCOUNT
  • REGION
  • USAGE_TYPE
  • OPERATION
  • INSTANCE_TYPE
  • PURCHASE_TYPE
  • RECORD_TYPE

Supported Metrics

  • UnblendedCost
  • AmortizedCost
  • NetAmortizedCost
  • NetUnblendedCost
  • NormalizedUsageAmount
  • UsageQuantity
  • BlendedCost

Security Considerations

  • This server requires read access to AWS Cost Explorer
  • Use least-privilege IAM policies
  • Store credentials securely using AWS best practices
  • Consider using IAM roles for EC2/Lambda deployments

Required AWS Permissions

Your AWS credentials need the following permissions:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "ce:GetCostAndUsage",
        "ce:GetDimensionValues"
      ],
      "Resource": "*"
    }
  ]
}

Troubleshooting

Common Issues

  1. "No credentials found"

    • Ensure AWS credentials are properly configured
    • Check that the profile name is correct (if using profiles)
  2. "Access denied"

    • Verify IAM permissions for Cost Explorer
    • Ensure billing access is enabled for your account
  3. "Invalid date format"

    • Use YYYY-MM-DD format for dates
    • Ensure end date is after start date

Testing

The server includes a test script to verify functionality:

# Basic functionality test (with credentials configured)
python test_server.py

Expected output when working correctly:

Initializing connection...
Available tools: ['get_cost_and_usage', 'get_dimension_values']
Testing get_cost_and_usage tool...

If you see "Unable to locate credentials", configure AWS credentials as described in the Configuration section above.

Development

To extend this server:

  1. Add new tools using the @server.tool() decorator
  2. Follow the MCP protocol specifications
  3. Use proper error handling for AWS API calls
  4. Test with different AWS account configurations

License

This project is open source. See LICENSE file for details.

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选