v0.1.1 – First Public Release
I'm excited to announce the first public release of AWS FinOps MCP Server, now available on PyPI.
This project brings AWS cost and resource visibility to your favourite LLM-powered environments using the Model Context Protocol (MCP) enabling natural-language FinOps queries from tools like Claude Desktop, Amazon Q CLI, and many other agents.
Features
- LLM-Ready FinOps: Integrates directly with Claude Desktop and other MCP clients to provide cost and audit analysis from your terminal or chat agent.
- Cost Explorer Insights: Query AWS cost and usage by service, tag, region, usage type, etc.
- FinOps Audit Tool: Identify stopped EC2s, unattached volumes, unused EIPs, instantly.
- Multi-Profile & Multi-Region Support: Works across all your AWS CLI profiles and multiple regions in one request.
- Secure by Design: Your AWS credentials never leave your machine. No backend, no token sharing.
Bug Fixes from v0.1.0 (unreleased)
- Fixed module import issue causing CLI execution to fail on fresh installs.
- Corrected packaging to ensure CLI command
aws-finops-mcp-serveris properly registered. - Updated dependency constraints and internal structure for PyPI compatibility.
Example Prompts
Claude Desktop
“Getusage-typebased cost for all my AWS CLI profiles for the last 90 days. And get theaudit reportfor all profiles for regionap-south-1. Make a finops report in a minimalistic and professional format.”
Amazon Q CLI
“Useaws-finopsMCP server to get monthly cost data forprofile 04for the last 6 months. Show trends by service.”
📘 Documentation
See the README for setup, configuration, sample prompts, IAM policies, and usage instructions.
🔗 Useful Links
- 🔗 PyPI: aws-finops-mcp-server
- 🔗 Live Sample Report: Comprehensive FinOps Report
- 🔗 Claude Desktop MCP Docs: claude.ai/download
- 🔗 Supported Clients: MCP Clients Directory