This is your comprehensive reference for managing EC2 instances, from basic launches to advanced patterns like mixed Spot and On-Demand Auto Scaling Groups. The skill triggers on practical signals like "spin up a server", SSH connection problems, or mentions of instance types (t3, m5, c7g), so it catches real-world scenarios even when you don't explicitly say "EC2". It covers the full lifecycle: launching instances with proper IMDSv2 configuration, creating AMIs, attaching IAM roles, setting up security groups, and cost optimization through Spot Instances and Graviton. The troubleshooting section is solid for debugging connectivity issues and the SSM Parameter Store pattern for AMI lookups is the right way to avoid hardcoded IDs.
- skills.sh installs
- 136
- GitHub stars
- 1.2K
- last updated
- Jun 8, 2026
Install
$ npx -y skills add itsmostafa/aws-agent-skills --skill ec2 --agent claude-codeInstalls into .claude/skills of the current project.
Files
AWS EC2
Amazon Elastic Compute Cloud (EC2) provides resizable compute capacity in the cloud.
Advanced patterns (Auto Scaling, Spot Fleets, Session Manager, Instance Connect, IMDS, Placement Groups, scheduled scaling): see instance-management.md.
Table of Contents
Core Concepts
Instance Types
| Category | Example | Use Case |
|---|---|---|
| General Purpose | t3, m6i, t4g (Graviton) | Web servers, dev environments |
| Compute Optimized | c6i, c7g (Graviton) | Batch processing, gaming |
| Memory Optimized | r6i, r7g (Graviton) | Databases, caching |
| Storage Optimized | i3, d3 | Data warehousing |
| Accelerated | p4d, g5 | ML, graphics |
Graviton (ARM) instances (t4g, m7g, c7g, r7g) are ~20% cheaper than x86 equivalents for the same performance — worth considering for new workloads.
Purchasing Options
Sponsors
Details
- Repository
- github.com/itsmostafa/aws-agent-skills
- Categories
- AI & Agent Building, Cloud & Infrastructure
- License
- MIT
- First seen
- Jun 11, 2026
