amazon-aurora-postgresql

Creates, sizes, and optimizes Amazon Aurora PostgreSQL clusters with cost analysis scripts.

Updated Jul 1, 2026
One-click install
npx skills add https://github.com/sakicodes/BuildFestHackathon26 --skill amazon-aurora-postgresql-sakicodes
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: amazon-aurora-postgresql
Source: https://github.com/sakicodes/BuildFestHackathon26/tree/main/.agents/skills/amazon-aurora-postgresql
Command: npx skills add https://github.com/sakicodes/BuildFestHackathon26 --skill amazon-aurora-postgresql-sakicodes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires boto3, and includes scripts (resource) and references (resource) components.

What problem does it solve? Managing Aurora PostgreSQL clusters involves complex decisions around provisioning, serverless ACU sizing, storage types, commitment pricing, and version upgrades, and mistakes can be costly or irreversible. This Skill routes each request to the right sub-skill, enforces safety guardrails on mutations, and runs deterministic pricing scripts instead of hand-calculating costs. ## Core Features & Use Cases - Cluster Creation with Express-First Routing: Provisions Aurora PostgreSQL via the single-API-call express configuration by default, falling back to full configuration only when VPC, custom KMS, or pinned engine versions are required. - Cost and Sizing Analysis: Bundled scripts compute Aurora serverless ACU sizing, I/O-Optimized breakeven points, and Reserved Instance vs Database Savings Plan comparisons with live AWS pricing or static fallbacks. - Safety Guardrails: Three-tier mutation policy confirms low-risk changes, demands risk briefings for high-impact changes, and blocks irreversible operations like cluster deletion with constructive alternatives. - Use Case: A user asks whether to switch their db.r7g.2xlarge cluster to Aurora serverless. The Skill loads the serverless-advisory reference, runs the ACU calculator with the user's CPU metrics, and returns a cost comparison table with a recommendation label. ## Quick Start Ask the assistant to create an Aurora PostgreSQL cluster or to estimate Aurora serverless ACU sizing and monthly cost for your current instance type and workload.

Frequently Asked Questions about amazon-aurora-postgresql

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create an Aurora PostgreSQL cluster quickly?

Use the express configuration, which provisions an Aurora PostgreSQL serverless cluster with a single create-db-cluster --with-express-configuration API call and no VPC setup. It uses IAM-only authentication via the Internet Access Gateway, and you connect with aws rds generate-db-auth-token.

How do I calculate Aurora serverless ACU sizing for my workload?

Run the bundled acu_calculator.py script with your instance type, CPU p95 and max percentages, and storage size. It recommends min and max ACU settings and compares serverless versus provisioned monthly costs using live AWS pricing or static fallbacks.

Reserved Instances vs Database Savings Plans for Aurora, which should I choose?

RIs offer deeper discounts up to roughly 66 percent on 3-year terms but lock you to one instance family, while DSP is family-agnostic, 1-year only, and the only option for Aurora serverless. Note that DSP does not cover older families like r6g or r5.

When should I switch to Aurora I/O-Optimized storage?

Switch when I/O charges exceed roughly 25 percent of your total cluster cost, which is the breakeven threshold against the 30 percent compute premium. The io_optimized_analyzer.py script computes the exact breakeven from your storage size and monthly I/O volume.

Can this skill delete or fail over my Aurora cluster?

No. Destructive and high-impact operations like delete-db-cluster, failover, major version upgrades, and public accessibility changes are blocked by design. The skill refuses, explains why, and offers a safe alternative such as a pre-deletion snapshot review or upgrade assessment.

Does the commitment pricing analysis work without AWS credentials?

Yes. The analyzer scripts support an offline mode using built-in us-east-1 pricing tables when credentials or the Pricing API are unavailable. You supply the instance type, instance count, region, or average ACU, and results are computed from static fallback rates.