Cloud Insight AI is a containerized cloud monitoring and analysis tool that processes cost and log data, generates AI-assisted insights, and visualizes the results on a web dashboard.
The application is controlled using a YAML configuration file.
This file defines:
- cost and log data sources
- error and cost thresholds
- feature toggles (AI insights, notifications)
- output and reporting settings
The backend runs inside a Docker container.
GitHub Actions builds the Docker image and pushes it to Amazon ECR.
Inside the container, analyzer.py acts as the entry point and coordinates:
- configuration loading
- cost analysis
- log processing
- optional AI-based insights
- report generation
Cost data from data/cost.json is analyzed to identify trends and anomalies.
Application, performance, and security logs are parsed to detect errors, warnings, and recurring issues.
AWS Comprehend is used to extract sentiment, key phrases, and entities from analysis summaries.
The analyzer generates:
final_report.json– frontend-ready analysis outputfinal_report.txt– human-readable summaryconfig.json– JSON representation of the active configuration
Generated JSON artifacts are uploaded to Amazon S3, which acts as a static data source for the dashboard.
A static dashboard fetches final_report.json from S3 and renders interactive visualizations.
├── README.md
├── _push.sh
├── architechture.png
├── cloud.png
├── config.json
├── config.yaml
├── data
│ ├── cost.json
│ ├── logs.txt
│ ├── performance-logs.txt
│ └── security-logs.txt
├── docker
│ └── Dockerfile
├── final_report.json
├── index.html
├── package-lock.json
├── requirements.txt
├── script.js
├── src
│ ├── analyzer.py
│ ├── comprehend_client.py
│ ├── config_loader.py
│ ├── cost_processor.py
│ ├── json_report_generator.py
│ ├── log_processor.py
│ └── notification_handler.py
└── style.css
- AWS S3 – Stores analysis reports and configuration artifacts
- Amazon ECR – Docker image registry
- Amazon ECS – Executes the containerized analyzer
- AWS IAM – Permissions for ECS, S3, ECR, and CI/CD
- AWS Comprehend – Optional NLP-based insight generation
- GitHub Actions – CI/CD pipeline
- Docker – Containerization
- Python – Backend analysis
- Chart.js – Interactive visualizations
- Frontend – Static HTML, CSS, JavaScript
Check out the full walkthrough and demo of Cloud Insight AI on YouTube:

