Comment by elC0mpa

Comment by elC0mpa 2 days ago

2 replies

Hi HN,

I’m a Cloud Architect, and I built aws-doctor because I found myself constantly running the same manual checks across different AWS accounts to find "zombie" resources. While AWS Trusted Advisor exists, the best checks are often locked behind paid Business/Enterprise support plans, and the AWS Console can be slow when you just want a quick "health check."

What it does: It’s a TUI (Terminal User Interface) that acts as a proactive checkup for your account.

Waste Detection: Scans for stopped instances (>30 days), unattached EBS volumes, unassociated Elastic IPs, and expiring Reserved Instances.

Cost Diagnosis: Compares your current month-to-date costs against the exact same period last month (e.g., Jan 1–15 vs. Feb 1–15) to spot spending velocity issues.

Trends: Visualizes cost history over the last 6 months.

The Tech Stack:

Written in Go (1.24).

Uses AWS SDK v2.

UI built with Charm's Bubbletea and Lipgloss (for the tables/styling).

It’s completely open-source and runs locally on your machine (using your standard ~/.aws/credentials).

I’d love to hear your feedback on the code structure or suggestions for other "waste patterns" I should add to the detection logic.

Thanks!

sandGorgon 2 days ago

have you considered using https://github.com/hashicorp/terraform-exec to run terraform inside you go process to manage the entire AWS connection piece. Terraform being largely rock solid and frequently updated on this.

could make this considerably more robust.

  • elC0mpa 17 hours ago

    Thanks a lot for your advice, I work every day with Terraform and I understand what you mean, but that is out of the scope at least right now, I want to keep this tool as simple as possible, but definitely this worth an analysis