zxkane aws-skills
COMMUNITYLABSCO SUMMARY
The six split as: aws-mcp-setup (the shared dependency, configuring AWS Documentation MCP access for the others), aws-cdk-development and aws-sst-development (two competing infrastructure-as-code paths, CDK or SST/Ion, each one telling an agent explicitly not to use the other's approach), aws-cost-operations (billing estimates, CloudWatch alarms, CloudTrail audits), aws-serverless-eda (Lambda, API Gateway, Step Functions, EventBridge patterns), and aws-agentic-ai (deploying and monitoring agents on Bedrock AgentCore's Gateway, Runtime, Memory, and Identity services). All six are mcp_service in our dependency check — none is a self-contained guide the way a pure-text skill would be.
This is for a team already building on AWS through Claude Code who wants per-domain expertise wired to the matching MCP servers — AWS Pricing, Cost Explorer, CloudWatch, the CDK MCP, Bedrock AgentCore's own services — rather than a general AWS reference. The README is explicit that aws-common (aws-mcp-setup) installs first as a dependency of the other plugins; skipping that order isn't documented as an option.
READ THE FULL ANALYSIS
The MCP server names are truncated to fit inside Bedrock's tool-name limit. The README's own naming scheme, mcp__plugin_{plugin}_{server}__{tool} with short IDs like cw for CloudWatch and sfn for Step Functions, exists because Bedrock caps a tool name at 64 characters — one sign of how many separate servers zxkane/aws-skills is wiring together under one set of plugins.
ALSO IN THIS PACKAGE
aws-common
Shared AWS agent skills including AWS Documentation MCP configuration for querying up-to-date AWS knowledge
aws-iac
AWS infrastructure-as-code skills covering both CDK (CloudFormation) and SST v4 (Pulumi/Ion): CDK best practices and Lambda workflows, plus SST resource authoring, linking, IAM scoping, deploy troubleshooting, and migration patterns
aws-cost-ops
AWS cost optimization, monitoring, and operational excellence with integrated MCP servers for billing, cost analysis, observability, and security assessment
serverless-eda
AWS serverless and event-driven architecture best practices based on Well-Architected Framework with MCP servers for SAM, Lambda, Step Functions, and messaging
aws-agentic-ai
AWS Bedrock AgentCore comprehensive expert for deploying and managing all AgentCore services including Gateway, Runtime, Memory, Identity, Code Interpreter, Browser, Observability, Agent Registry, and Evaluations
WHAT'S INSIDE
6 showing · 6 totalaws-agentic-ai
Amazon's platform for running AI agents in production is nine separate services; this points you at the right one for the job and shows how they fit together.
aws-cdk-development
Describe your AWS setup in TypeScript or Python instead of clicking through a console, and catch the security and cost problems before it goes live.
aws-cost-operations
How much AWS is costing you, and whether the things running on it are actually healthy.
aws-mcp-setup
Two ways to plug an AI assistant into AWS: one that only reads the documentation and needs no account, and one that can call AWS for real.
aws-serverless-eda
Building an app out of small pieces of code that run only when something happens — and the rules of thumb that keep it from falling over at scale.
aws-sst-development
SST is a way of writing your AWS setup as TypeScript code; this carries the house conventions and the specific mistakes that took real projects down.
HOW TO GET IT
npx skills add zxkane/aws-skillsnpx skills add zxkane/aws-skills --skill <name> --full-depthPick the skill name from the Skills tab — each entry there installs independently.
/plugin marketplace add zxkane/aws-skills/plugin install aws-common@aws-skills/plugin install aws-iac@aws-skills/plugin install aws-cost-ops@aws-skillsTyped inside the agent's own prompt, not in a terminal. The marketplace lists 5 plugins; three are shown. Every one installs the same way, with its own name before @aws-skills.