AWS account-to-account migration toolkit: 9 specialist skills + 9 sub-agents + 4 deterministic Workflow orchestrators. Produces schema-validated JSON artifacts at every stage. Terraform-portable HCL output with parameterized account IDs, regions, and AZs.
Compare source and target AWS accounts after the cutover. Re-runs the same describe-* scope against both profiles, structurally diffs, categorizes drift (missing/extra/config/security/cost/scope), and emits audit-diff.json + audit-report.md.
Discover-only pass — inventory the source account + dependency analysis, but skip Terraform generation and migration planning. Use to understand what is in the source account before committing to a migration.
Execute an already-approved cutover-checklist.md against the target account, one resource at a time, with mandatory per-step human approval and resumable journal. Halts and offers rollback on failure. Polls long-running data-plane jobs.
Run the full AWS account-to-account migration pipeline end-to-end (Discover → Generate → Cutover). Halts at the cutover checklist; human runs the actual cutover, then invokes /aws-migration-architect:audit.
Generate the control-plane cutover checklist — 7 phases (Globals → Networking → Storage Containers → Database Containers → Compute Containers → DNS Scaffolding → Control Plane Validation) of Terraform module applies and AWS control-plane API operations. NO data movement, NO writable-data freeze, NO production DNS swap. Output is cutover-checklist-control-plane.json + .md. Use when invoked by the cutover-control-plane skill or the migrate workflow.
Generate the data-plane cutover checklist — 5 phases (Pre-Staging → Bulk Transfers → Application Data → Cutover → Data Validation) that move data, freeze writes, swap DNS, promote replicas, and validate the copy. Consumes data-migration-plan.json as primary input. NO terraform applies, NO IAM creation. Output is cutover-checklist-data-plane.json + .md. Use when invoked by the cutover-data-plane skill or the migrate workflow.
Actually execute the cutover one resource at a time. Reads BOTH cutover checklists (cutover-checklist-control-plane.json AND cutover-checklist-data-plane.json) plus data-migration-plan.json + migration-plan.json + dependency-graph.json. Compiles execution-steps.json walking control-plane first then data-plane (preview/execute/verify/rollback/poll per step). Walks the list with mandatory per-step human approval. Halts and offers rollback on failure. Resumable via append-only JSONL journal (re-verifies in-flight steps against AWS before continuing). Polls long-running data-plane jobs (DataSync, DMS, S3 Batch, DynamoDB export/import). Use when invoked by the cutover-executor skill or the migrate workflow's Execute phase.
Plan data movement for every data-bearing resource. Sizes each datastore via AWS APIs (CloudWatch + describe), picks transfer tool + mode per the rules in the data-migration-planner skill (size, RPO, encryption), estimates wall-clock time using per-tool throughput models, prices the transfer via the awspricing MCP, applies criticality-tier RPO/RTO defaults, computes freeze windows for non-continuous strategies, defines validation methods, and emits data-migration-plan.json + .md. Use when invoked by the data-migration-planner skill or the migrate workflow's DataPlan phase.
Read-only AWS dependency analysis sub-agent. Reads inventory.json and walks every resource configuration to enumerate cross-resource references (SG, IAM, Lambda env, S3 policy, Route53, etc.), classifies IAM trusts (cross-account/OIDC/IRSA/SAML), detects hard-coded values (account IDs, regions, EIPs, ARNs, domains), assigns Low/Medium/High risk per resource, and emits Mermaid architecture diagrams. Use when invoked by the dependency-analyzer skill or the migrate workflow.
Generate the control-plane cutover runbook — the steps that create the empty target shape (IAM, networking, KMS keys, empty resource containers, DNS scaffolding) via Terraform module applies and AWS control-plane API calls. Phase 0 (Globals: IAM/Route53 root/CloudFront/Backup) → 1 (Networking) → 2 (Storage containers) → 3 (Database containers) → 4 (Compute containers) → 5 (DNS scaffolding, no record changes) → 6 (Control plane validation). Produces cutover-checklist-control-plane.md + .json. Runs BEFORE the data-plane runbook.
Generate the data-plane cutover runbook — the steps that move actual data (snapshot shares, KMS grants, AMI shares, DataSync, DMS, S3 sync, snapshot-restore, ECR push, secret values), freeze writes during cutover, swap DNS / promote replicas, and validate the copy. Consumes data-migration-plan.json for sizing, strategy, freeze windows, and validation criteria. Phase 1 (Pre-Staging) → 2 (Bulk Transfers) → 3 (Application Data) → 4 (Cutover: freeze + promote + swap) → 5 (Data Validation). Produces cutover-checklist-data-plane.md + .json. Runs AFTER the control-plane runbook completes.
Execute BOTH approved cutover checklists (control plane then data plane) against the target AWS account, one resource at a time. Reads cutover-checklist-control-plane.json, cutover-checklist-data-plane.json, data-migration-plan.json, migration-plan.json. Builds an execution-steps.json with preview/execute/verify/rollback/poll per step. Walks control-plane steps first (Terraform module applies + AWS control-plane API), then data-plane steps (snapshot share, restore, DataSync, DMS, freeze, route53 swap, validation). Mandatory per-step human approval. Halts and offers rollback on failure. Resumable via append-only JSONL journal — on resume, re-verifies any in-flight steps against AWS before continuing.
Plan the data movement for every data-bearing resource in scope. Sizes each datastore via AWS APIs, picks the right transfer tool + mode (bulk vs bulk+delta vs continuous), estimates wall-clock transfer time and dollar cost (egress, cross-region, cross-account, tool runtime, double storage), captures encryption requirements (KMS grants, re-encryption), surfaces RPO/RTO targets per criticality tier, and produces freeze windows and validation criteria per datastore. Output is data-migration-plan.json + .md — consumed by cutover-data-plane to inject real timings into the cutover checklist.
Find the hidden coupling that breaks migrations. Walks resource-to-resource references (SG rules, Lambda env vars, S3 policies, IAM trust chains), classifies IAM trusts (cross-account, OIDC, IRSA, SAML), detects hard-coded values (account IDs, regions, EIPs, ARNs, domains), assigns Low/Medium/High risk per resource, and emits Mermaid architecture diagrams. Use after `inventory` and before `terraform-generator` or `migration-planner`.
External network access
Connects to servers outside your machine
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A Claude Code plugin that turns the messy reality of AWS account-to-account migration into a deterministic, schema-validated pipeline. Nine specialist skills, nine bounded sub-agents, four slash commands, and an orchestrator that takes you from a source-account scan all the way through resource-by-resource execution against the target. The cutover is split into two distinct runbooks — control plane (Terraform + AWS API to create the empty target shape) and data plane (snapshot share, DataSync, DMS, freeze, swap, validate) — each separately approved by a human.
This is the plugin you point at a source AWS account when you need to move it to a different AWS account. It is not a generic AWS expert; it is a focused migration toolkit.
| Step | Skill | What you get |
|---|---|---|
| 1 | inventory | inventory.json — every resource in the source account, scoped by region + service + tag |
| 2 | dependency-analyzer | dependency-graph.json, IAM trust classification, hard-coded value detection, per-resource risk score, 4 Mermaid architecture diagrams |
| 3 | terraform-generator | terraform/{networking,compute,storage,databases,iam,dns}/ — re-deployable HCL with account IDs / regions / AZs parameterized |
| 4 | migration-planner | cost-baseline.json → readiness-score.json → migration-plan.json + .md (6 phases with rollback) |
| 5 | data-migration-planner | data-migration-plan.json + .md — per-datastore sizing (CloudWatch + describe APIs), transfer tool + mode by size/RPO/encryption, wall-clock time estimate with confidence, transfer cost (egress + tool runtime + double-storage) via awspricing MCP, RPO/RTO targets per criticality tier, freeze windows for non-continuous strategies, validation methods + acceptance criteria, rollback retention |
| 6 | cutover-control-plane | cutover-checklist-control-plane.md + .json — 7-phase runbook (0 Globals → 1 Networking → 2 Storage Containers → 3 Database Containers → 4 Compute Containers → 5 DNS Scaffolding → 6 Control Plane Validation). Terraform module applies + AWS control-plane API only. NO data movement, NO freeze, NO production DNS swap. Hands off to data-plane runbook via handoff_to_data_plane.criteria[]. |
| 7 | cutover-data-plane | cutover-checklist-data-plane.md + .json — 5-phase runbook (1 Pre-Staging → 2 Bulk Transfers → 3 Application Data → 4 Cutover (freeze + promote + swap) → 5 Data Validation). Consumes data-migration-plan.json for sizing/strategy/freeze-windows/validation. Marks irreversible steps (route53 swap, DMS promote). Aggregates freeze_windows[]. |
| 8 | cutover-executor | execution-steps.json, execution-log.jsonl, execution-report.md — walks BOTH checklists in order (control plane then data plane) with mandatory per-step human approval, polls long-running data-plane jobs (DataSync/DMS/S3 Batch/DynamoDB export-import), halts and offers rollback on failure, resumable via append-only journal. Refuses to advance from control plane to data plane until operator confirms handoff criteria + data-plane IAM is attached. |
| 9 | post-migration-auditor | audit-diff.json + audit-report.md — verifies parity between source and target after cutover |
Two modes:
/aws-migration-architect:migrate to take you from inventory through the printed checklist. Read and sign the checklist (APPROVED BY: <name> ON: <date> line near the top). Then /aws-migration-architect:execute --run-id <id> walks the checklist resource-by-resource against the target with per-step approval. Then /aws-migration-architect:audit --run-id <id> verifies parity.In Claude Code:
/plugin marketplace add /Users/pventrapragada/Desktop/workspace/aws_migration/aws-migration-architect
/plugin install aws-migration-architect
You should now see:
inventory, dependency-analyzer, terraform-generator, migration-planner, data-migration-planner, cutover-control-plane, cutover-data-plane, cutover-executor, post-migration-auditor/aws-migration-architect:migrate, :discover, :execute, :auditawsknowledge, awsiac, awspricingVerify with /plugin.
Install on your workstation (not in any AWS account):
# macOS (Homebrew)
brew install awscli terraform uv
# Linux (one option of several)
pipx install awscli # or use the AWS-published installer
brew install terraform || download from terraform.io
curl -LsSf https://astral.sh/uv/install.sh | sh
npx claudepluginhub pavankishore1987/aws-migration-architect --plugin aws-migration-architectHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.