CostQ — Cloud Cost Optimization with Amazon Q

English | 日本語

CostQ is an open-source project that demonstrates how to use Amazon Q to analyze AWS bills and generate actionable cost optimization reports. The project provides Chinese/English/Japanese prompt templates, report templates, and execution handbooks, helping you quickly understand cloud cost structures and optimization opportunities with Amazon Q.

Why CostQ

• Results-Oriented: Executive summary → Optimization plan → Execution checklist.
• Tri-lingual Support: Chinese / English / Japanese prompts and reports.
• Plug-and-Play Data: Integrates with APIs such as Cost Explorer, Cost Optimization Hub, Compute Optimizer, SP/RI coverage & utilization, S3/EC2 monitoring metrics.

Repository Structure

.
├─ prompts/       # Prompt templates for different roles
│  ├─ zh-CN/     
│  ├─ en/
│  └─ ja/
├─ reports/       # Cost analysis report templates
│  ├─ report.zh-CN.md   
│  ├─ report.en.md
│  └─ report.ja.md
├─ handbook/      # Amazon Q installation and execution handbook
│  ├─ runbook.zh-CN.md  
│  ├─ runbook.en.md
│  └─ runbook.ja.md

Prerequisites

• Cost Explorer enabled.
• Amazon Q CLI installed and configured.

Quick Start

1. Prepare Amazon Q: Follow handbook/runbook.zh-CN.md to create permissions and install Amazon Q.
2. Choose a Language: Use the prompts/zh-CN prompt template in Amazon Q.
3. Generate Reports: Ask Q to produce:
   • Executive Report: Cost trends, cost drivers, savings opportunities.

Prompt Content Overview

• Multi-dimensional Cost Overview: Cost efficiency metrics + cost composition.
• Intelligent Anomaly Detection & Root Cause Analysis: Anomaly triggers + CloudTrail deep analysis.
• Optimization Strategies Based on Cost Analysis: Identify optimization potential + recommendations.
• Actionable Cost Optimization Reports

Suggested Amazon Q Usage

• First generate an executive summary report → Then ask Q to cite data evidence for each conclusion.
• Next generate a cost optimization report → Require owners, risk/rollback, and validation metrics.

Security & Privacy

• Do not submit raw bills, account IDs, or other sensitive information.
• Always anonymize before external sharing.
• This project has no official affiliation with AWS, for learning and collaboration only.

Contributing

PRs are welcome! Please:

• Use Conventional Commits (e.g., feat:, fix:).
• Add sample/test data (anonymized).
• Keep prompts auditable and reproducible.

License

MIT