feat: add aws-dev-toolkit plugin (v0.12.0 — 34 skills, 11 agents, 3 MCP servers)

[rsmets]

Summary

aws-dev-toolkit is a comprehensive AWS development toolkit plugin for Claude Code — 34 skills, 11 specialized agents, and 3 MCP servers in a single plugin for building, migrating, and reviewing well-architected applications on AWS.

This is the implementation accompanying RFC #108.

Why a unified plugin?

See the detailed argument on the RFC, but in short:

• More coverage in 1 plugin than all 7 existing plugins combined — 34 skills vs 21, 11 agents vs 0, 19 AWS services with dedicated skills vs ~8
• Cross-cutting concerns (security review, cost estimation, Well-Architected alignment) have no natural home in fragmented plugins
• Workflow composition — aws-plan chains discovery → design → security review → cost estimate in one flow
• Zero MCP duplication — current plugins duplicate awsknowledge across 3 separate plugins
• Industry precedent and Anthropic's own research support integrated harnesses over narrow tools

Plugin structure (v0.12.0)

plugins/aws-dev-toolkit/
├── .claude-plugin/plugin.json
├── .mcp.json                    # 3 MCP servers
├── skills/                      # 34 skills (each with SKILL.md + references/)
│   ├── Service-specific (19): agentcore, api-gateway, bedrock, cloudfront,
│   │   dynamodb, ec2, ecs, eks, iam, iot, lambda, messaging, mlops,
│   │   networking, observability, rds-aurora, s3, security-review, step-functions
│   └── Workflow/cross-cutting (15): aws-architect, aws-compare, aws-debug,
│       aws-diagram, aws-health-check, aws-migrate, aws-plan, azure-to-aws,
│       challenger, cost-check, customer-ideation, gcp-to-aws, iac-scaffold,
│       strands-agent, well-architected
├── agents/                      # 11 specialized agents
│   ├── agentcore-sme, aws-explorer, bedrock-sme, container-sme,
│   │   cost-optimizer, iac-reviewer, migration-advisor, networking-sme,
│   │   observability-sme, serverless-sme, well-architected-reviewer

Skills inventory (34 skills)

Service-specific skills (19):

Skill	Description
agentcore	Bedrock AgentCore platform — Runtime, Memory, Gateway, Identity, Policy, multi-agent architectures
api-gateway	Amazon API Gateway (REST, HTTP, WebSocket APIs), authorizers, CORS, throttling
bedrock	Bedrock model selection, agents, guardrails, fine-tuning, cost modeling
cloudfront	CloudFront distributions, cache behaviors, edge functions, OAC
dynamodb	Table design, access patterns, GSI/LSI strategies, single-table design
ec2	Instance selection, auto-scaling, Spot instances, storage optimization
ecs	Fargate vs EC2 launch types, task definitions, deployment strategies
eks	Managed node groups, Fargate profiles, IRSA, Karpenter autoscaling
iam	Policies, roles, permission boundaries, SCPs, Identity Center, Access Analyzer
iot	IoT Core MQTT/shadows/rules, Greengrass v2, SiteWise, fleet provisioning, device security
lambda	Function design, cold starts, Powertools, concurrency, event sources, SnapStart
messaging	SQS, SNS, EventBridge — event-driven architectures, fan-out, DLQ patterns
mlops	SageMaker training/inference/pipelines, MLflow, model monitoring, cost optimization
networking	VPC architecture, subnets, security groups, Transit Gateway, Route53
observability	CloudWatch metrics, logs, alarms, dashboards, X-Ray tracing
rds-aurora	Engine selection, Aurora vs RDS, Serverless v2, HA, RDS Proxy, blue/green
s3	Storage strategies, bucket policies, lifecycle policies, performance
security-review	Review infrastructure code and configurations for security issues
step-functions	Workflows — saga patterns, parallel tasks, error recovery, direct integrations

Workflow & cross-cutting skills (15):

Skill	Description
aws-architect	Design and review AWS architectures following Well-Architected Framework
aws-compare	Side-by-side architecture comparison across cost, complexity, performance
aws-debug	Debug infrastructure issues, deployment failures, and runtime errors
aws-diagram	Mermaid + ASCII architecture diagrams from descriptions or IaC
aws-health-check	5-minute account scan for top security/cost/reliability issues
aws-migrate	Migration assessment: discovery, service mapping, wave planning, cost estimation
aws-plan	End-to-end discovery → design → security review → cost estimate
azure-to-aws	Azure to AWS migration guidance with service mappings and assessment
challenger	Adversarial reviewer that stress-tests recommendations
cost-check	Analyze and optimize AWS costs, investigate unexpected charges
customer-ideation	Guide customers from idea to AWS architecture
gcp-to-aws	GCP to AWS migration guidance with service mappings and assessment
iac-scaffold	Scaffold IaC projects (CDK, Terraform, SAM, CloudFormation)
strands-agent	Scaffold AI agents using Strands SDK with Bedrock AgentCore
well-architected	Formal Well-Architected Framework reviews against six pillars

33 reference files

Skills include references/*.md files with detailed CLI commands, code examples, policy templates, and pipeline recipes — keeping SKILL.md focused on guidance while providing deep-dive examples when needed.

Quality standards

• Every opinionated recommendation includes a concrete justification — cost numbers, failure modes, operational tradeoffs
• Related Skills cross-references across 14+ skills
• Anti-patterns with consequences on every service skill
• Output format tables standardized across skills

Test plan

• Install plugin via claude plugins add ./plugins/aws-dev-toolkit
• Verify all 34 skills appear in / slash command list
• Test aws-plan end-to-end workflow (discovery → design → security → cost)
• Test aws-health-check against a live AWS account
• Verify MCP servers connect (awsiac, awsknowledge, awspricing)
• Spawn agent teams (e.g., cost-optimizer + security-review composition)
• Verify reference files load correctly for skills with references/ directories