aws-cost-explorer-mcp-server

概述

此 MCP 服务器使用户能够通过自然语言界面查询和分析 AWS 支出数据。它连接到 AWS Cost Explorer API 和 CloudWatch Logs 以获取 Bedrock 模型调用数据。该服务器支持本地（stdio）和远程（SSE）部署，包含详细的 Claude Desktop 集成配置说明和可选的 Chainlit Web 界面。在适当配置的情况下，它还提供跨账户 AWS 支出数据访问功能。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• get_ec2_spend_last_dayget_ec2_spend_last_day()

  Retrieves EC2 spending data for the previous day

  试试:What was my EC2 spend yesterday?试试:Show me my EC2 costs from the last day
• get_detailed_breakdown_by_dayget_detailed_breakdown_by_day(days=7)

  Delivers a comprehensive analysis of costs by region, service, and instance type

  试试:Show me my AWS costs broken down by day for the last week试试:Analyze my spending by region and service for the past 7 days
• get_bedrock_daily_usage_statsget_bedrock_daily_usage_stats(days=7, region='us-east-1', log_group_name='BedrockModelInvocationLogGroup')

  Delivers a per-day breakdown of model usage by region and users

  试试:What was my Bedrock usage like in the last week?试试:Show me daily Bedrock usage stats for us-west-2 region
• get_bedrock_hourly_usage_statsget_bedrock_hourly_usage_stats(days=7, region='us-east-1', log_group_name='BedrockModelInvocationLogGroup')

  Delivers a per-day per-hour breakdown of model usage by region and users

  试试:Show me hourly Bedrock usage patterns for the last 7 days试试:What were my peak usage hours for Bedrock last week?

可对比工具

安装

# macOS 和 Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

git clone https://github.com/aarora79/aws-cost-explorer-mcp.git
cd aws-cost-explorer-mcp
uv venv --python 3.12 && source .venv/bin/activate && uv pip install --requirement pyproject.toml

{
  "mcpServers": {
    "aws-cost-explorer": {
      "command": "uv",
      "args": [
          "--directory",
          "/path/to/aws-cost-explorer-mcp",
          "run",
          "server.py"
      ],
      "env": {
        "AWS_ACCESS_KEY_ID": "YOUR_ACCESS_KEY_ID",
        "AWS_SECRET_ACCESS_KEY": "YOUR_SECRET_ACCESS_KEY",
        "AWS_REGION": "us-east-1",
        "BEDROCK_LOG_GROUP_NAME": "YOUR_CLOUDWATCH_BEDROCK_LOG_GROUP_NAME",
        "MCP_TRANSPORT": "stdio"
      }
    }
  }
}

FAQ

这个 MCP 服务器可以访问其他 AWS 账户的支出数据吗？

是的，您可以在启动服务器时设置 CROSS_ACCOUNT_ROLE_NAME 参数以承担其他 AWS 账户中的角色。

服务器提供了哪些工具？

该服务器提供了 EC2 支出分析、Bedrock 使用统计以及按日期、区域、服务和实例类型进行的详细成本分解工具。

aws-cost-explorer-mcp-server 对比