What are the prerequisites for using AWS FinOps MCP Server?
You need Python 3.10+, the AWS CLI installed and configured, at least one AWS profile with read‑only IAM permissions, and an MCP‑compatible client (e.g., Claude Desktop).
@ravikiranvm
Basic information
An MCP (Model Context Protocol) server that brings powerful AWS FinOps capabilities directly into your AI assistant. Analyze cloud costs, audit for waste, and get budget insights using natural language, all while keeping your credentials secure on your local machine.
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"aws-finops-mcp-server": {
"command": "python",
"args": [
"-m",
"pip",
"install",
"--user",
"pipx"
]
}
}
}Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
An MCP (Model Context Protocol) server that brings AWS FinOps capabilities into your AI assistant. It analyzes cloud costs, audits for waste, and provides budget insights using natural language, while keeping your credentials secure on your local machine.
Install via pipx install aws-finops-mcp-server (or using uv). Configure one or more AWS CLI profiles. Then add the server to your MCP client’s configuration file (e.g., Claude for Desktop’s claude_desktop_config.json). After restarting the client, ask questions about your AWS finances in plain English.
You need Python 3.10+, the AWS CLI installed and configured, at least one AWS profile with read‑only IAM permissions, and an MCP‑compatible client (e.g., Claude Desktop).
Each `get_cost` call makes 2 calls to the AWS Cost Explorer API, costing $0.01. The `run_finops_audit` tool uses other AWS APIs (EC2, Budgets) that are generally in the free tier or have negligible cost.
Claude Desktop and Amazon Q CLI are explicitly supported. Any MCP‑compatible client that supports tools can also be used.
The IAM policy must allow the following read‑only actions: `ce:GetCostAndUsage`, `budgets:ViewBudget`, `ec2:DescribeInstances`, `ec2:DescribeVolumes`, `ec2:DescribeAddresses`, and `sts:GetCallerIdentity`.
The server runs locally and uses your existing AWS CLI configuration. Credentials never leave your machine; API calls are made directly from your local environment to AWS.
MCP server for Backblaze B2 Cloud Storage: a focused, safe 40-tool surface (17 native B2 SDK, 19 S3 data-plane, 4 analytics) for any MCP-compatible AI client, currently incubating in Backblaze-Labs

HostTracker is a website and infrastructure monitoring service that has run external checks since 2004 for more than 500,000 websites. Its official MCP server lets an AI assistant operate a HostTracker account directly i

Read-only, governed infrastructure context for any AI. OneIQ gives Claude, ChatGPT, Copilot and other MCP clients answers from your VMware, Nutanix, Kubernetes and cloud environments, without access to production systems
Model Context Protocol (MCP) server for Kubernetes and OpenShift
Comments