[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7A3DP74DDHAP9QXGFNVCQAE":3,"skill-same:01M7A3DP74DDHAP9QXGFNVCQAE":146},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":42,"commitSha":43,"skillDirectory":44,"forksCount":45,"repositoryHtmlUrl":46,"repositoryHomepageUrl":6,"ownerHtmlUrl":47,"ownerBlogUrl":6,"ownerDescription":48,"updatedAt":49,"install":50,"resources":63},"01M7A3DP74DDHAP9QXGFNVCQAE","appmesh-virtual-service-deployer",null,"Provisions AWS App Mesh virtual service layer resources with production defaults: service mesh creation (egress filter DROP_ALL vs ALLOW_ALL), virtual nodes with DNS service discovery (hostname-based) vs Cloud Map service discovery (namespace-based, auto-registration), virtual services backed by virtual nodes or virtual routers, virtual routers with weighted route configuration (HTTP \u002F TCP \u002F gRPC routes for canary traffic shifting), timeout and retry policies (per-route and per-virtual-node), circuit breaker via connection pool limits and outlier detection, virtual gateway for north-south ingress (ALB\u002FNLB → gateway → mesh), Envoy proxy injection (auto-inject on EKS via App Mesh mutating admission webhook, ECS sidecar registration, EC2 manual sidecar), xDS protocol (Envoy fetches. Triggers: create app mesh virtual service, app mesh virtual node, app mesh virtual router, weighted routing canary, app mesh envoy inject, app mesh mTLS, virtual gateway ingress, app mesh cloud map discovery, app mesh dns discovery.","jhkchan\u002Faws-agent-skills","jhkchan","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1083902?v=4","01M7A2SVQYP9GQTPTDXP5JKYQ4","Jacky Chan","01M7A2SVS117YWCTZABB75HZR1",4,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: appmesh-virtual-service-deployer\ndescription: 'Provisions AWS App Mesh virtual service layer resources with production defaults: service mesh creation (egress filter DROP_ALL vs ALLOW_ALL), virtual nodes with DNS service discovery (hostname-based) vs Cloud Map service discovery (namespace-based, auto-registration), virtual services backed by virtual nodes or virtual routers, virtual routers with weighted route configuration (HTTP \u002F TCP \u002F gRPC routes for canary traffic shifting), timeout and retry policies (per-route and per-virtual-node), circuit breaker via connection pool limits and outlier detection, virtual gateway for north-south ingress (ALB\u002FNLB → gateway → mesh), Envoy proxy injection (auto-inject on EKS via App Mesh mutating admission webhook, ECS sidecar registration, EC2 manual sidecar), xDS protocol (Envoy fetches. Triggers: create app mesh virtual service, app mesh virtual node, app mesh virtual router, weighted routing canary, app mesh envoy inject, app mesh mTLS, virtual gateway ingress, app mesh cloud map discovery, app mesh dns discovery.'\nlicense: Apache-2.0\ncompatibility: 'Agent runtime that reads SKILL.md (Claude Code, Cursor, Windsurf, Codex, Gemini). For live deployment: AWS CLI v2 with appmesh create-mesh, create-virtual-node, create-virtual-router, create-route, create-virtual-service, create-virtual-gateway, create-gateway-route (AWS CLI v2, SSO or key-based credentials). For EKS Envoy injection: kubectl and the App Mesh Controller with mutating webhook configuration. For Cloud Map: servicediscovery create-service, create-http-namespace. For mTLS:...'\nmetadata:\n  domain: aws-cloudops\n  complexity: high\n  requires_llm: 'true'\n  phase: '1'\n  supports_pipeline: 'true'\n  entry_point: 'false'\n  family: Networking\n  task_type: deploy\n  skill_class: capability\n  lifecycle_status: active\n  verdict_shape: READY_TO_DEPLOY | PREREQUISITES_MISSING\n  version: 0.1.0\n  author: Jacky Chan — AWS Community Builder\n  tags: aws, appmesh, service-mesh, virtual-service, envoy, networking, deploy, cloudops, weighted-routing, canary, cloud-map, mtls, circuit-breaking\n  dependencies: aws-orchestrator\n  keywords: aws, app mesh, service mesh, virtual node, virtual service, virtual router, virtual gateway, weighted routing, canary deploy, traffic shifting, envoy proxy, envoy sidecar, auto inject, mutating webhook, xds protocol, cloud map, dns service discovery, service discovery, circuit breaker, outlier detection, connection pool, retry policy, timeout policy, mTLS, acm private ca, x-ray tracing, cloudwatch metrics, cloudops, deploy, networking\n  when_to_use: Invoke when the user wants to create an App Mesh service mesh with virtual nodes, virtual services, and virtual routers, configure weighted routing for canary traffic shifting, set up a virtual gateway for north-south ingress, inject Envoy sidecars (auto-inject on EKS via webhook, ECS sidecar, or EC2 manual), enable mTLS via ACM Private CA, configure circuit breakers and outlier detection, or integrate CloudWatch metrics and X-Ray tracing. Do NOT invoke for AWS Cloud Map service discovery without App Mesh (use servicediscovery skills), Istio service mesh (different control plane), or AWS Gateway Load Balancer (different Layer 3\u002F4 inspection service).\n---\n\n# App Mesh Virtual Service Deployer\n\nAn AWS CloudOps agent skill that provisions AWS App Mesh virtual\nservice layer resources with correct defaults. The skill walks the\noperator through service mesh creation, virtual node service discovery\n(DNS vs Cloud Map), virtual router weighted route configuration for\ncanary deployments, Envoy sidecar injection (auto-inject on EKS via\nmutating webhook), virtual gateway ingress, mTLS via ACM Private CA,\ncircuit breaking, and observability (CloudWatch + X-Ray). It captures\nthe mesh topology decisions, explains why each default matters, and\nemits a READY_TO_DEPLOY checklist with copy-pasteable verification\ncommands.\n\n## Activation keywords\n\ncreate app mesh virtual service, app mesh virtual node, app mesh\nvirtual router, weighted routing canary, app mesh envoy inject, app\nmesh mTLS, virtual gateway ingress, app mesh cloud map discovery, app\nmesh dns discovery.\n\n## STRICT output contract\n\nWhen this skill is invoked with an App Mesh virtual-service-\nprovisioning request (create a mesh, configure virtual nodes,\nweighted routing, virtual gateway, Envoy injection, mTLS, circuit\nbreaking, or a partial configuration), the agent MUST respond with\nthe READY_TO_DEPLOY checklist defined in the \"Output format\" section\nusing the literal all-caps labels `APP_MESH:`, `VERDICT:`,\n`CHECKLIST:`, and `VERIFICATION_COMMANDS:`. Do NOT preface the\nchecklist with prose, headings, or disclaimers — emit the block as\nthe first lines of the response. This contract is what assertion-\nbased evals and downstream provisioning pipelines rely on; deviating\nfrom the literal labels breaks automation silently.\n\nIf any prerequisite is missing, the verdict is `PREREQUISITES_MISSING`\nwith a specific gap citation in the checklist (marked `[✗]`), and\n`READY_TO_DEPLOY` MUST NOT also appear.\n\n## Quick navigation\n\n| Section | When to read |\n|---|---|\n| Prerequisites | Always — verify before provisioning |\n| Step 1 — Service mesh creation | Mesh + egress filter |\n| Step 2 — Virtual nodes (DNS vs Cloud Map) | Service discovery decision |\n| Step 3 — Virtual routers and weighted routes | Canary traffic shifting |\n| Step 4 — Route policies (timeout, retry) | Resilience configuration |\n| Step 5 — Circuit breaker and outlier detection | Fault isolation |\n| Step 6 — Virtual services | Abstraction layer |\n| Step 7 — Virtual gateway (north-south ingress) | External traffic entry |\n| Step 8 — Envoy sidecar injection | Data plane provisioning |\n| Step 9 — mTLS via ACM Private CA | East-west security |\n| Step 10 — Observability (CloudWatch + X-Ray) | Monitoring and tracing |\n| Step 11 — Mesh scope (namespace vs cluster) | EKS scoping |\n| Step 12 — xDS protocol | Control plane communication |\n| Step 13 — Recent features | Latest |\n| NEVER do these things | Review before signing off |\n| Output format | The literal checklist template |\n| references\u002Fservice-discovery-and-routing.md | DNS vs Cloud Map + routing detail |\n| references\u002Fenvoy-and-mtls-guide.md | Envoy injection + mTLS detail |\n\n## Mindset\n\n**One-line takeaway:** An App Mesh virtual service is an abstraction\nover a real service. A virtual node represents a deployable unit\n(with DNS or Cloud Map service discovery). A virtual router directs\ntraffic to virtual nodes via weighted routes. Envoy sidecars (auto-\ninjected on EKS via a mutating webhook) enforce the routing, retry,\ntimeout, and mTLS policies. The control plane (App Mesh) pushes\nconfiguration to Envoy via the xDS protocol.\n\nThree misconceptions dominate App Mesh virtual service misdesign at\nprovisioning time:\n\n- **\"Virtual nodes and virtual services are the same.\"** They are\n  NOT. A virtual node represents a concrete deployable unit (a\n  Kubernetes Deployment, an ECS service, an EC2 fleet) with a service\n  discovery mechanism (DNS hostname or Cloud Map service). A virtual\n  service is an abstraction that clients call — it is backed by either\n  a virtual node (direct routing, no traffic splitting) or a virtual\n  router (weighted routing for canary\u002Fblue-green). The virtual service\n  is the DNS name clients use; the virtual node is where traffic\n  actually goes.\n\n- **\"Weighted routing works without a virtual router.\"** It does NOT.\n  Traffic splitting (e.g., 90% to v1, 10% to v2 for canary) requires\n  a virtual router with a weighted HTTP\u002FTCP\u002FgRPC route. If the\n  virtual service points directly to a virtual node (no router), ALL\n  traffic goes to that single node — no splitting is possible. This\n  is the #1 cause of \"my canary isn't working\" — the virtual service\n  has no router.\n\n- **\"Envoy sidecars are automatically injected on all platforms.\"**\n  Only on EKS with the App Mesh Controller installed and the namespace\n  labeled for injection. The controller installs a mutating admission\n  webhook that injects the Envoy container into pods. On ECS, you\n  must add the Envoy sidecar to the task definition manually. On EC2,\n  you run Envoy as a process. Without the sidecar, mesh policies are\n  NOT enforced — traffic flows directly, bypassing routing rules,\n  retries, and mTLS.\n\n## Configuration dependency graph (novel heuristic)\n\nApp Mesh virtual service configurations are NOT independent. The mesh\nmust exist before virtual nodes. Virtual nodes must exist before\nvirtual routers (routes reference them as targets). The virtual\nservice must reference either a node or a router. Envoy sidecars\nmust be injected for policies to take effect. Use this graph to\nsequence provisioning.\n\n| Configuration | Hard dependencies (API error without) | Silent failure \u002F immutability | Enables downstream |\n|---|---|---|---|\n| Service mesh | IAM appmesh:CreateMesh; unique mesh name | Egress filter DROP_ALL blocks all non-mesh egress by default | mesh container |\n| Virtual node (DNS) | Mesh exists; DNS hostname resolvable | DNS hostname must point to the real service; no auto-registration | node with DNS discovery |\n| Virtual node (Cloud Map) | Mesh exists; Cloud Map namespace + service exist | Cloud Map service must register instances; empty service = no endpoints | node with Cloud Map discovery (auto-registration) |\n| Virtual router | Mesh exists | Router without routes = no traffic flows (virtual service backed by router gets 503) | routing layer |\n| Weighted route | Router exists; two+ virtual nodes exist as targets | Weights must sum to 100; single-node route = all traffic to one node (no canary) | canary \u002F blue-green traffic shifting |\n| Virtual service | Mesh exists; backed by virtual node OR virtual router | Service pointing to a node (not router) = no traffic splitting possible | client-facing abstraction |\n| Virtual gateway | Mesh exists; gateway listener configured | Gateway without gateway routes = no external traffic enters mesh | north-south ingress |\n| Envoy sidecar (EKS) | App Mesh Controller installed; namespace labeled | Pods without sidecar = policies NOT enforced (traffic bypasses mesh) | policy enforcement |\n| Envoy sidecar (ECS) | Envoy container in task definition; App Mesh proxy configuration | Missing Envoy = mesh policies silently bypassed | policy enforcement |\n| mTLS (listener TLS) | ACM Private CA ACTIVE; virtual node\u002Fgateway listener TLS configured | mTLS in STRICT mode without valid client cert = all connections rejected | east-west\u002Fnorth-south encryption |\n| mTLS (peer TLS) | ACM Private CA ACTIVE; SDS backend configured | Peer mTLS PERMISSIVE mode allows non-mTLS (migration); STRICT blocks | backpressure encryption |\n| Circuit breaker | Virtual node backend defaults (connection pool, outlier detection) | Connection pool too low = healthy endpoints ejected; too high = no protection | fault isolation |\n\n**The weighted-routing-requires-router row is the one a baseline\nmodel misses.** Creating a virtual service that points directly to a\nvirtual node means NO traffic splitting is possible. The virtual\nrouter with weighted routes is the ONLY mechanism for canary\u002Fblue-\ngreen. The Envoy-sidecar-required row is the second commonly\nmisunderstood step.\n\n**Cross-dependency gotchas:**\nCross-dependency gotchas moved verbatim to [references\u002Fadvanced-patterns.md](references\u002Fadvanced-patterns.md).\nLoad on demand when sequencing mesh, node, router, gateway, or mTLS provisioning.\n\n## Expert heuristic: DNS vs Cloud Map service discovery\n\nA baseline model says \"pick any service discovery.\" The correct\nheuristic recognizes that DNS and Cloud Map serve different use cases,\nand the wrong choice causes stale backends or registration failures.\n\n```text\nService discovery decision:\n  ├── DNS service discovery\n  │     → Virtual node specifies a DNS hostname (e.g., my-service.local)\n  │     → Envoy resolves the hostname to get backend IPs\n  │     → DNS TTL controls staleness (high TTL = stale endpoints)\n  │     → Best for: services with stable IPs, ALB\u002FNLB fronted services\n  │     → No registration needed (DNS already resolves)\n  │     → Limitation: DNS caching can delay backend removal\n  │\n  └── Cloud Map service discovery\n        → Virtual node specifies a Cloud Map namespace + service name\n        → Instances self-register with Cloud Map on startup\n        → Envoy queries Cloud Map API for live backends\n        → Best for: dynamic scaling (ECS, EKS, EC2 ASG)\n        → Auto-registration: instances register\u002Fderegister on start\u002Fstop\n        → Lower latency for endpoint changes (no DNS cache)\n        → Requires: Cloud Map namespace + service created beforehand\n```\n\n**Key implication:** Cloud Map is the right choice for dynamic workloads\n(ECS tasks, EKS pods that scale). DNS is simpler but stale for rapidly\nchanging backends. The virtual node's `serviceDiscovery` field\ndetermines which mechanism is used — it CANNOT be both.\n\n## Expert heuristic: weighted routing for canary deploy\n\nA baseline model deploys v2 and hopes traffic shifts. The correct\nheuristic recognizes that canary traffic shifting requires a virtual\nrouter with weighted HTTP routes.\n\n```text\nCanary deploy flow with weighted routing:\n  1. Virtual service \"checkout.mesh.local\" → virtual router\n  2. Virtual router has HTTP route \"checkout-route\"\n  3. Route initially: checkout-v1 (weight 100), checkout-v2 (weight 0)\n  4. Deploy v2 (new virtual node \"checkout-v2\")\n  5. Update route: checkout-v1 (weight 90), checkout-v2 (weight 10)\n     → 10% of traffic goes to v2 (canary)\n  6. Monitor metrics (error rate, latency) on v2\n  7. If healthy: update to checkout-v1 (weight 50), checkout-v2 (weight 50)\n  8. If healthy: update to checkout-v1 (weight 0), checkout-v2 (weight 100)\n  9. v2 is now live; v1 can be scaled down\n\n  WITHOUT a virtual router (virtual service → virtual node directly):\n    → ALL traffic goes to the single node\n    → NO canary possible (no traffic splitting mechanism)\n    → Must swap the node's backend (instant cutover, no gradual shift)\n```\n\n**Key implication:** the virtual router is the ONLY mechanism for\nweighted traffic splitting. A virtual service backed by a virtual\nnode (no router) cannot do canary. This is the #1 cause of \"my canary\nisn't working.\"\n\n## Expert heuristic: Envoy sidecar auto-inject via webhook\n\nPlatform-by-platform injection detail moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nLoad on demand when deciding EKS auto-inject vs ECS sidecar vs EC2 manual Envoy.\n\n## Prerequisites (verify before provisioning)\n\nBefore emitting provisioning commands, verify these prerequisites. If\nany are missing, the verdict is **PREREQUISITES_MISSING**.\n\n| Prerequisite | Why it matters | How to verify |\n|---|---|---|\n| Service mesh name (unique in region) | Mesh names must be unique | `aws appmesh list-meshes` |\n| Service discovery decision (DNS vs Cloud Map) | Determines virtual node config | Assess workload type |\n| Cloud Map namespace exists (if Cloud Map) | Virtual node references it | `aws servicediscovery list-namespaces` |\n| Cloud Map service exists (if Cloud Map) | Virtual node references it | `aws servicediscovery list-services` |\n| DNS hostname resolvable (if DNS discovery) | Envoy resolves it for backends | `dig \u003Chostname>` or `nslookup` |\n| EKS cluster running (if EKS) | Envoy auto-inject needs cluster | `kubectl get nodes` |\n| App Mesh Controller installed (if EKS) | Provides the mutating webhook | `kubectl get pods -n appmesh-system` |\n| Namespace labeled for injection (if EKS auto-inject) | Webhook only fires on labeled namespaces | `kubectl get ns \u003Cns> --show-labels` |\n| ACM Private CA ACTIVE (if mTLS) | Issues Envoy certificates | `aws acm-pca list-certificate-authorities` |\n| IAM appmesh:Create* permissions | Required for all create calls | Check IAM policy |\n| Virtual node backend protocol (HTTP\u002FTCP\u002FgRPC) | Route type must match | Confirm protocol |\n\nIf any prerequisite is missing, output `VERDICT: PREREQUISITES_MISSING`\nand cite the specific gap.\n\n## Step 1 — Service mesh creation\n\nThe service mesh is the top-level container for all virtual nodes,\nrouters, services, and gateways.\n\n**Create the mesh:**\n\n```bash\naws appmesh create-mesh \\\n  --mesh-name production-mesh \\\n  --spec '{\"egressFilter\":{\"type\":\"ALLOW_ALL\"}}' \\\n  --region us-east-1\n```\n\n**Egress filter options:**\n\n| Filter | Behavior |\n|---|---|\n| ALLOW_ALL | Pods can reach any destination (mesh and non-mesh) |\n| DROP_ALL | Pods can only reach mesh virtual services (all other egress blocked) |\n\n**Critical:** DROP_ALL is more secure but breaks non-mesh dependencies\n(database endpoints, external APIs). Start with ALLOW_ALL, then tighten\nto DROP_ALL with explicit virtual services for allowed egress.\n\n## Step 2 — Virtual nodes (DNS vs Cloud Map)\n\ncreate-virtual-node CLI for both discovery modes moved verbatim to [references\u002Fservice-discovery-and-routing.md](references\u002Fservice-discovery-and-routing.md).\nLoad on demand when emitting virtual-node provisioning commands.\n\n## Step 3 — Virtual routers and weighted routes\n\ncreate-virtual-router and weighted create-route CLI moved verbatim to [references\u002Fservice-discovery-and-routing.md](references\u002Fservice-discovery-and-routing.md).\nLoad on demand when configuring canary traffic splitting.\n\n## Step 4 — Route policies (timeout, retry)\n\nRetry\u002Ftimeout route-policy CLI and retry event types moved verbatim to [references\u002Fservice-discovery-and-routing.md](references\u002Fservice-discovery-and-routing.md).\nLoad on demand when adding resilience policies to a route.\n\n## Step 5 — Circuit breaker and outlier detection\n\nCircuit-breaker and outlier-detection payloads moved verbatim to [references\u002Fworked-examples.md](references\u002Fworked-examples.md).\nLoad on demand when configuring backendDefaults health checks or outlierDetection.\n\n## Step 6 — Virtual services\n\nA virtual service is the client-facing abstraction. It is backed by\neither a virtual node (direct, no splitting) or a virtual router\n(weighted routing).\n\n**Virtual service backed by a virtual router (canary-capable):**\n\n```bash\naws appmesh create-virtual-service \\\n  --mesh-name production-mesh \\\n  --virtual-service-name checkout.mesh.local \\\n  --spec '{\n    \"provider\": {\n      \"virtualRouter\": {\n        \"virtualRouterName\": \"checkout-router\"\n      }\n    }\n  }' \\\n  --region us-east-1\n```\n\n**Virtual service backed by a virtual node (no splitting):**\n\n```bash\naws appmesh create-virtual-service \\\n  --mesh-name production-mesh \\\n  --virtual-service-name inventory.mesh.local \\\n  --spec '{\n    \"provider\": {\n      \"virtualNode\": {\n        \"virtualNodeName\": \"inventory-v1\"\n      }\n    }\n  }' \\\n  --region us-east-1\n```\n\n**Critical:** a virtual service backed by a virtual node CANNOT do\nweighted routing. Only a virtual service backed by a virtual router\ncan split traffic. Choose the provider based on whether canary\u002Fblue-\ngreen is needed.\n\n## Step 7 — Virtual gateway (north-south ingress)\n\ncreate-virtual-gateway and create-gateway-route CLI moved verbatim to [references\u002Fworked-examples.md](references\u002Fworked-examples.md).\nLoad on demand when provisioning north-south ingress.\n\n## Step 8 — Envoy sidecar injection\n\nEKS webhook install, ECS sidecar JSON, and the critical no-Envoy failure mode moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nLoad on demand when injecting Envoy on any platform.\n\n## Step 9 — mTLS via ACM Private CA\n\nListener TLS and backend peer TLS (SDS) payloads moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nLoad on demand when enabling STRICT or PERMISSIVE mTLS.\n\n## Step 10 — Observability (CloudWatch + X-Ray)\n\nAccess-log payload and Envoy metric names moved verbatim to [references\u002Fworked-examples.md](references\u002Fworked-examples.md).\nLoad on demand when wiring observability.\n\n## Step 11 — Mesh scope (namespace vs cluster)\n\nNamespace-scoping kubectl commands moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nLoad on demand when scoping injection on EKS.\n\n## Step 12 — xDS protocol\n\nxDS flow walkthrough moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nKey implication kept here: route weight changes need no pod restarts — xDS pushes within seconds.\n\n## Step 13 — Recent features\n\nRecent AWS features (2023-2026) moved verbatim to [references\u002Fadvanced-patterns.md](references\u002Fadvanced-patterns.md).\nLoad on demand when evaluating newest App Mesh capabilities.\n\n## NEVER do these things\n\n1. **NEVER back a virtual service with a virtual node if you need\n   canary routing.** Only a virtual router can do weighted traffic\n   splitting. A virtual service backed by a virtual node sends ALL\n   traffic to that single node.\n\n2. **NEVER assume Envoy is auto-injected on all platforms.** Auto-\n   inject works ONLY on EKS with the App Mesh Controller and labeled\n   namespaces. ECS and EC2 require manual sidecar configuration.\n\n3. **NEVER use mTLS STRICT mode during migration.** STRICT rejects\n   all connections without valid certificates. Start with PERMISSIVE\n   (allows mixing), verify all services have certs, then switch to\n   STRICT.\n\n4. **NEVER forget the Cloud Map namespace and service.** Cloud Map\n   service discovery requires a pre-existing namespace and service.\n   Without them, the virtual node has no backends and returns 503.\n\n5. **NEVER use DROP_ALL egress filter without planning.** DROP_ALL\n   blocks all non-mesh egress. If services depend on databases,\n   external APIs, or S3, these connections will break unless explicit\n   virtual services are created for allowed egress.\n\n6. **NEVER assume route weight changes require pod restarts.** The\n   xDS protocol pushes configuration changes in near-real-time via\n   streaming. Weight changes take effect within seconds without\n   restarts.\n\n7. **NEVER confuse virtual node service discovery types.** DNS and\n   Cloud Map are mutually exclusive — a virtual node uses ONE or the\n   other. Mixing them is not supported.\n\n8. **NEVER forget that Envoy resolves the DNS hostname at runtime.**\n   DNS service discovery relies on the hostname being resolvable by\n   Envoy. If DNS is misconfigured, Envoy has no backends.\n\n9. **NEVER create a virtual gateway without gateway routes.** The\n   virtual gateway is the listener; gateway routes direct traffic to\n   virtual services. Without routes, no external traffic enters the\n   mesh.\n\n10. **NEVER set circuit breaker connection pool limits too low.**\n    Low limits eject healthy endpoints under normal load. Start with\n    generous limits and tune based on observed behavior.\n\n## Output format\n\n```text\nAPP_MESH: \u003Cmesh-name> — virtual service \u003Cvs-name> (\u003Cprotocol>)\nVERDICT: READY_TO_DEPLOY | PREREQUISITES_MISSING\nCHECKLIST:\n  [✓|✗] Service mesh: \u003Cmesh-name> (egress: ALLOW_ALL | DROP_ALL)\n  [✓|✗] Virtual node: \u003Cvn-name> (service discovery: DNS \u003Chostname> | Cloud Map \u003Cnamespace>\u002F\u003Cservice>)\n  [✓|✗] Virtual router: \u003Cvr-name> (listener: \u003Cport>\u002F\u003Cprotocol>)\n  [✓|✗] Weighted route: \u003Croute-name> — \u003Cvn-1> (\u003Cweight>%), \u003Cvn-2> (\u003Cweight>%)\n  [✓|✗] Virtual service: \u003Cvs-name> → backed by \u003Cvirtual-router | virtual-node>\n  [✓|✗] Retry policy: maxRetries=\u003Cn>, perRetry=\u003Cms>, events=\u003Cevent-list>\n  [✓|✗] Timeout: request=\u003Cvalue>\u003Cunit>\n  [✓|✗] Circuit breaker: healthCheck path=\u003Cpath>, outlier detection maxServerErrors=\u003Cn>\n  [✓|✗] Virtual gateway: \u003Cvgw-name> (listener: \u003Cport>\u002F\u003Cprotocol>)\n  [✓|✗] Gateway route: \u003Cgr-name> → \u003Cvirtual-service-name>\n  [✓|✗] Envoy sidecar: auto-inject (EKS namespace=\u003Cns>) | manual (ECS task | EC2)\n  [✓|✗] mTLS: STRICT | PERMISSIVE | disabled (ACM Private CA \u003Cca-arn>)\n  [✓|✗] Observability: CloudWatch metrics enabled, X-Ray tracing enabled\n  [✓|✗] Mesh ARN: arn:aws:appmesh:\u003Cregion>:\u003Cacct>:mesh\u002F\u003Cmesh-name>\nVERIFICATION_COMMANDS:\n  aws appmesh describe-mesh --mesh-name \u003Cmesh-name> --region \u003Cregion>\n  aws appmesh describe-virtual-node --mesh-name \u003Cmesh-name> --virtual-node-name \u003Cvn-name> --region \u003Cregion>\n  aws appmesh describe-route --mesh-name \u003Cmesh-name> --virtual-router-name \u003Cvr-name> --route-name \u003Croute-name> --region \u003Cregion>\n  aws appmesh describe-virtual-service --mesh-name \u003Cmesh-name> --virtual-service-name \u003Cvs-name> --region \u003Cregion>\n```\n\n### Worked example — weighted canary with Cloud Map discovery and Envoy auto-inject\n\n```text\nAPP_MESH: production-mesh — virtual service checkout.mesh.local (http)\nVERDICT: READY_TO_DEPLOY\nCHECKLIST:\n  [✓] Service mesh: production-mesh (egress: ALLOW_ALL)\n  [✓] Virtual node: checkout-v1 (service discovery: Cloud Map mesh-services\u002Fcheckout)\n  [✓] Virtual node: checkout-v2 (service discovery: Cloud Map mesh-services\u002Fcheckout-v2)\n  [✓] Virtual router: checkout-router (listener: 8080\u002Fhttp)\n  [✓] Weighted route: checkout-canary — checkout-v1 (90%), checkout-v2 (10%)\n  [✓] Virtual service: checkout.mesh.local → backed by virtual-router checkout-router\n  [✓] Retry policy: maxRetries=3, perRetry=2000ms, events=server-error,gateway-error\n  [✓] Timeout: request=15s\n  [✓] Circuit breaker: healthCheck path=\u002Fhealth, outlier detection maxServerErrors=5\n  [✓] Envoy sidecar: auto-inject (EKS namespace=default, webhook=appmesh-controller)\n  [✓] mTLS: PERMISSIVE (migration mode, ACM Private CA active)\n  [✓] Observability: CloudWatch metrics enabled, X-Ray tracing enabled\n  [✓] Mesh ARN: arn:aws:appmesh:us-east-1:123456789012:mesh\u002Fproduction-mesh\nVERIFICATION_COMMANDS:\n  aws appmesh describe-mesh --mesh-name production-mesh --region us-east-1\n  aws appmesh describe-virtual-node --mesh-name production-mesh --virtual-node-name checkout-v1 --region us-east-1\n  aws appmesh describe-route --mesh-name production-mesh --virtual-router-name checkout-router --route-name checkout-canary --region us-east-1\n  aws appmesh describe-virtual-service --mesh-name production-mesh --virtual-service-name checkout.mesh.local --region us-east-1\n```\n\n## Error handling\n\nError-handling deep dives (503s, canary not splitting, Envoy not injected, mTLS rejected, DROP_ALL breakage, empty Cloud Map) moved verbatim to [references\u002Ferror-handling.md](references\u002Ferror-handling.md).\nLoad on demand when a deployment or runtime failure must be diagnosed.\n\n## References (load on demand)\n\n- [references\u002Fservice-discovery-and-routing.md](references\u002Fservice-discovery-and-routing.md) — DNS vs Cloud Map and routing deep dive; now also holds Steps 2-4 (virtual nodes, weighted routes, route policies) moved from SKILL.md.\n- [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md) — Envoy injection and mTLS deep dive; now also holds the Envoy auto-inject heuristic and Steps 8, 9, 11, 12 moved from SKILL.md.\n- [references\u002Fworked-examples.md](references\u002Fworked-examples.md) — secondary worked examples moved from SKILL.md: circuit breaker (Step 5), virtual gateway (Step 7), observability (Step 10).\n- [references\u002Ferror-handling.md](references\u002Ferror-handling.md) — error-handling deep dives moved from SKILL.md: 503 no healthy upstream, canary not splitting, Envoy not injected, mTLS rejections, DROP_ALL egress breakage, Cloud Map with no instances.\n- [references\u002Fadvanced-patterns.md](references\u002Fadvanced-patterns.md) — cross-dependency gotchas and Recent AWS features 2023-2026 (Step 13) moved from SKILL.md.\n\n## Domain\n\nAWS CloudOps \u002F AWS App Mesh Virtual Service Layer Provisioning &\nService Mesh Networking.\n\n## AWS documentation\n\n- **App Mesh User Guide** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fwelcome.html\n- **Virtual nodes** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fvirtual_nodes.html\n- **Virtual routers and routes** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fvirtual_routers.html\n- **Virtual services** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fvirtual_services.html\n- **Virtual gateways** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fvirtual_gateways.html\n- **Envoy proxy** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fenvoy.html\n- **mTLS** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fmtls.html\n- **Cloud Map service discovery** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fcloudmap.html\n- **App Mesh Controller for EKS** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fgetting-started-kubernetes.html\n- **CloudWatch metrics** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fmetrics.html\n","# App Mesh Virtual Service Deployer\n\nAn AWS CloudOps agent skill that provisions AWS App Mesh virtual\nservice layer resources with correct defaults. The skill walks the\noperator through service mesh creation, virtual node service discovery\n(DNS vs Cloud Map), virtual router weighted route configuration for\ncanary deployments, Envoy sidecar injection (auto-inject on EKS via\nmutating webhook), virtual gateway ingress, mTLS via ACM Private CA,\ncircuit breaking, and observability (CloudWatch + X-Ray). It captures\nthe mesh topology decisions, explains why each default matters, and\nemits a READY_TO_DEPLOY checklist with copy-pasteable verification\ncommands.\n\n## Activation keywords\n\ncreate app mesh virtual service, app mesh virtual node, app mesh\nvirtual router, weighted routing canary, app mesh envoy inject, app\nmesh mTLS, virtual gateway ingress, app mesh cloud map discovery, app\nmesh dns discovery.\n\n## STRICT output contract\n\nWhen this skill is invoked with an App Mesh virtual-service-\nprovisioning request (create a mesh, configure virtual nodes,\nweighted routing, virtual gateway, Envoy injection, mTLS, circuit\nbreaking, or a partial configuration), the agent MUST respond with\nthe READY_TO_DEPLOY checklist defined in the \"Output format\" section\nusing the literal all-caps labels `APP_MESH:`, `VERDICT:`,\n`CHECKLIST:`, and `VERIFICATION_COMMANDS:`. Do NOT preface the\nchecklist with prose, headings, or disclaimers — emit the block as\nthe first lines of the response. This contract is what assertion-\nbased evals and downstream provisioning pipelines rely on; deviating\nfrom the literal labels breaks automation silently.\n\nIf any prerequisite is missing, the verdict is `PREREQUISITES_MISSING`\nwith a specific gap citation in the checklist (marked `[✗]`), and\n`READY_TO_DEPLOY` MUST NOT also appear.\n\n## Quick navigation\n\n| Section | When to read |\n|---|---|\n| Prerequisites | Always — verify before provisioning |\n| Step 1 — Service mesh creation | Mesh + egress filter |\n| Step 2 — Virtual nodes (DNS vs Cloud Map) | Service discovery decision |\n| Step 3 — Virtual routers and weighted routes | Canary traffic shifting |\n| Step 4 — Route policies (timeout, retry) | Resilience configuration |\n| Step 5 — Circuit breaker and outlier detection | Fault isolation |\n| Step 6 — Virtual services | Abstraction layer |\n| Step 7 — Virtual gateway (north-south ingress) | External traffic entry |\n| Step 8 — Envoy sidecar injection | Data plane provisioning |\n| Step 9 — mTLS via ACM Private CA | East-west security |\n| Step 10 — Observability (CloudWatch + X-Ray) | Monitoring and tracing |\n| Step 11 — Mesh scope (namespace vs cluster) | EKS scoping |\n| Step 12 — xDS protocol | Control plane communication |\n| Step 13 — Recent features | Latest |\n| NEVER do these things | Review before signing off |\n| Output format | The literal checklist template |\n| references\u002Fservice-discovery-and-routing.md | DNS vs Cloud Map + routing detail |\n| references\u002Fenvoy-and-mtls-guide.md | Envoy injection + mTLS detail |\n\n## Mindset\n\n**One-line takeaway:** An App Mesh virtual service is an abstraction\nover a real service. A virtual node represents a deployable unit\n(with DNS or Cloud Map service discovery). A virtual router directs\ntraffic to virtual nodes via weighted routes. Envoy sidecars (auto-\ninjected on EKS via a mutating webhook) enforce the routing, retry,\ntimeout, and mTLS policies. The control plane (App Mesh) pushes\nconfiguration to Envoy via the xDS protocol.\n\nThree misconceptions dominate App Mesh virtual service misdesign at\nprovisioning time:\n\n- **\"Virtual nodes and virtual services are the same.\"** They are\n  NOT. A virtual node represents a concrete deployable unit (a\n  Kubernetes Deployment, an ECS service, an EC2 fleet) with a service\n  discovery mechanism (DNS hostname or Cloud Map service). A virtual\n  service is an abstraction that clients call — it is backed by either\n  a virtual node (direct routing, no traffic splitting) or a virtual\n  router (weighted routing for canary\u002Fblue-green). The virtual service\n  is the DNS name clients use; the virtual node is where traffic\n  actually goes.\n\n- **\"Weighted routing works without a virtual router.\"** It does NOT.\n  Traffic splitting (e.g., 90% to v1, 10% to v2 for canary) requires\n  a virtual router with a weighted HTTP\u002FTCP\u002FgRPC route. If the\n  virtual service points directly to a virtual node (no router), ALL\n  traffic goes to that single node — no splitting is possible. This\n  is the #1 cause of \"my canary isn't working\" — the virtual service\n  has no router.\n\n- **\"Envoy sidecars are automatically injected on all platforms.\"**\n  Only on EKS with the App Mesh Controller installed and the namespace\n  labeled for injection. The controller installs a mutating admission\n  webhook that injects the Envoy container into pods. On ECS, you\n  must add the Envoy sidecar to the task definition manually. On EC2,\n  you run Envoy as a process. Without the sidecar, mesh policies are\n  NOT enforced — traffic flows directly, bypassing routing rules,\n  retries, and mTLS.\n\n## Configuration dependency graph (novel heuristic)\n\nApp Mesh virtual service configurations are NOT independent. The mesh\nmust exist before virtual nodes. Virtual nodes must exist before\nvirtual routers (routes reference them as targets). The virtual\nservice must reference either a node or a router. Envoy sidecars\nmust be injected for policies to take effect. Use this graph to\nsequence provisioning.\n\n| Configuration | Hard dependencies (API error without) | Silent failure \u002F immutability | Enables downstream |\n|---|---|---|---|\n| Service mesh | IAM appmesh:CreateMesh; unique mesh name | Egress filter DROP_ALL blocks all non-mesh egress by default | mesh container |\n| Virtual node (DNS) | Mesh exists; DNS hostname resolvable | DNS hostname must point to the real service; no auto-registration | node with DNS discovery |\n| Virtual node (Cloud Map) | Mesh exists; Cloud Map namespace + service exist | Cloud Map service must register instances; empty service = no endpoints | node with Cloud Map discovery (auto-registration) |\n| Virtual router | Mesh exists | Router without routes = no traffic flows (virtual service backed by router gets 503) | routing layer |\n| Weighted route | Router exists; two+ virtual nodes exist as targets | Weights must sum to 100; single-node route = all traffic to one node (no canary) | canary \u002F blue-green traffic shifting |\n| Virtual service | Mesh exists; backed by virtual node OR virtual router | Service pointing to a node (not router) = no traffic splitting possible | client-facing abstraction |\n| Virtual gateway | Mesh exists; gateway listener configured | Gateway without gateway routes = no external traffic enters mesh | north-south ingress |\n| Envoy sidecar (EKS) | App Mesh Controller installed; namespace labeled | Pods without sidecar = policies NOT enforced (traffic bypasses mesh) | policy enforcement |\n| Envoy sidecar (ECS) | Envoy container in task definition; App Mesh proxy configuration | Missing Envoy = mesh policies silently bypassed | policy enforcement |\n| mTLS (listener TLS) | ACM Private CA ACTIVE; virtual node\u002Fgateway listener TLS configured | mTLS in STRICT mode without valid client cert = all connections rejected | east-west\u002Fnorth-south encryption |\n| mTLS (peer TLS) | ACM Private CA ACTIVE; SDS backend configured | Peer mTLS PERMISSIVE mode allows non-mTLS (migration); STRICT blocks | backpressure encryption |\n| Circuit breaker | Virtual node backend defaults (connection pool, outlier detection) | Connection pool too low = healthy endpoints ejected; too high = no protection | fault isolation |\n\n**The weighted-routing-requires-router row is the one a baseline\nmodel misses.** Creating a virtual service that points directly to a\nvirtual node means NO traffic splitting is possible. The virtual\nrouter with weighted routes is the ONLY mechanism for canary\u002Fblue-\ngreen. The Envoy-sidecar-required row is the second commonly\nmisunderstood step.\n\n**Cross-dependency gotchas:**\nCross-dependency gotchas moved verbatim to [references\u002Fadvanced-patterns.md](references\u002Fadvanced-patterns.md).\nLoad on demand when sequencing mesh, node, router, gateway, or mTLS provisioning.\n\n## Expert heuristic: DNS vs Cloud Map service discovery\n\nA baseline model says \"pick any service discovery.\" The correct\nheuristic recognizes that DNS and Cloud Map serve different use cases,\nand the wrong choice causes stale backends or registration failures.\n\n```text\nService discovery decision:\n  ├── DNS service discovery\n  │     → Virtual node specifies a DNS hostname (e.g., my-service.local)\n  │     → Envoy resolves the hostname to get backend IPs\n  │     → DNS TTL controls staleness (high TTL = stale endpoints)\n  │     → Best for: services with stable IPs, ALB\u002FNLB fronted services\n  │     → No registration needed (DNS already resolves)\n  │     → Limitation: DNS caching can delay backend removal\n  │\n  └── Cloud Map service discovery\n        → Virtual node specifies a Cloud Map namespace + service name\n        → Instances self-register with Cloud Map on startup\n        → Envoy queries Cloud Map API for live backends\n        → Best for: dynamic scaling (ECS, EKS, EC2 ASG)\n        → Auto-registration: instances register\u002Fderegister on start\u002Fstop\n        → Lower latency for endpoint changes (no DNS cache)\n        → Requires: Cloud Map namespace + service created beforehand\n```\n\n**Key implication:** Cloud Map is the right choice for dynamic workloads\n(ECS tasks, EKS pods that scale). DNS is simpler but stale for rapidly\nchanging backends. The virtual node's `serviceDiscovery` field\ndetermines which mechanism is used — it CANNOT be both.\n\n## Expert heuristic: weighted routing for canary deploy\n\nA baseline model deploys v2 and hopes traffic shifts. The correct\nheuristic recognizes that canary traffic shifting requires a virtual\nrouter with weighted HTTP routes.\n\n```text\nCanary deploy flow with weighted routing:\n  1. Virtual service \"checkout.mesh.local\" → virtual router\n  2. Virtual router has HTTP route \"checkout-route\"\n  3. Route initially: checkout-v1 (weight 100), checkout-v2 (weight 0)\n  4. Deploy v2 (new virtual node \"checkout-v2\")\n  5. Update route: checkout-v1 (weight 90), checkout-v2 (weight 10)\n     → 10% of traffic goes to v2 (canary)\n  6. Monitor metrics (error rate, latency) on v2\n  7. If healthy: update to checkout-v1 (weight 50), checkout-v2 (weight 50)\n  8. If healthy: update to checkout-v1 (weight 0), checkout-v2 (weight 100)\n  9. v2 is now live; v1 can be scaled down\n\n  WITHOUT a virtual router (virtual service → virtual node directly):\n    → ALL traffic goes to the single node\n    → NO canary possible (no traffic splitting mechanism)\n    → Must swap the node's backend (instant cutover, no gradual shift)\n```\n\n**Key implication:** the virtual router is the ONLY mechanism for\nweighted traffic splitting. A virtual service backed by a virtual\nnode (no router) cannot do canary. This is the #1 cause of \"my canary\nisn't working.\"\n\n## Expert heuristic: Envoy sidecar auto-inject via webhook\n\nPlatform-by-platform injection detail moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nLoad on demand when deciding EKS auto-inject vs ECS sidecar vs EC2 manual Envoy.\n\n## Prerequisites (verify before provisioning)\n\nBefore emitting provisioning commands, verify these prerequisites. If\nany are missing, the verdict is **PREREQUISITES_MISSING**.\n\n| Prerequisite | Why it matters | How to verify |\n|---|---|---|\n| Service mesh name (unique in region) | Mesh names must be unique | `aws appmesh list-meshes` |\n| Service discovery decision (DNS vs Cloud Map) | Determines virtual node config | Assess workload type |\n| Cloud Map namespace exists (if Cloud Map) | Virtual node references it | `aws servicediscovery list-namespaces` |\n| Cloud Map service exists (if Cloud Map) | Virtual node references it | `aws servicediscovery list-services` |\n| DNS hostname resolvable (if DNS discovery) | Envoy resolves it for backends | `dig \u003Chostname>` or `nslookup` |\n| EKS cluster running (if EKS) | Envoy auto-inject needs cluster | `kubectl get nodes` |\n| App Mesh Controller installed (if EKS) | Provides the mutating webhook | `kubectl get pods -n appmesh-system` |\n| Namespace labeled for injection (if EKS auto-inject) | Webhook only fires on labeled namespaces | `kubectl get ns \u003Cns> --show-labels` |\n| ACM Private CA ACTIVE (if mTLS) | Issues Envoy certificates | `aws acm-pca list-certificate-authorities` |\n| IAM appmesh:Create* permissions | Required for all create calls | Check IAM policy |\n| Virtual node backend protocol (HTTP\u002FTCP\u002FgRPC) | Route type must match | Confirm protocol |\n\nIf any prerequisite is missing, output `VERDICT: PREREQUISITES_MISSING`\nand cite the specific gap.\n\n## Step 1 — Service mesh creation\n\nThe service mesh is the top-level container for all virtual nodes,\nrouters, services, and gateways.\n\n**Create the mesh:**\n\n```bash\naws appmesh create-mesh \\\n  --mesh-name production-mesh \\\n  --spec '{\"egressFilter\":{\"type\":\"ALLOW_ALL\"}}' \\\n  --region us-east-1\n```\n\n**Egress filter options:**\n\n| Filter | Behavior |\n|---|---|\n| ALLOW_ALL | Pods can reach any destination (mesh and non-mesh) |\n| DROP_ALL | Pods can only reach mesh virtual services (all other egress blocked) |\n\n**Critical:** DROP_ALL is more secure but breaks non-mesh dependencies\n(database endpoints, external APIs). Start with ALLOW_ALL, then tighten\nto DROP_ALL with explicit virtual services for allowed egress.\n\n## Step 2 — Virtual nodes (DNS vs Cloud Map)\n\ncreate-virtual-node CLI for both discovery modes moved verbatim to [references\u002Fservice-discovery-and-routing.md](references\u002Fservice-discovery-and-routing.md).\nLoad on demand when emitting virtual-node provisioning commands.\n\n## Step 3 — Virtual routers and weighted routes\n\ncreate-virtual-router and weighted create-route CLI moved verbatim to [references\u002Fservice-discovery-and-routing.md](references\u002Fservice-discovery-and-routing.md).\nLoad on demand when configuring canary traffic splitting.\n\n## Step 4 — Route policies (timeout, retry)\n\nRetry\u002Ftimeout route-policy CLI and retry event types moved verbatim to [references\u002Fservice-discovery-and-routing.md](references\u002Fservice-discovery-and-routing.md).\nLoad on demand when adding resilience policies to a route.\n\n## Step 5 — Circuit breaker and outlier detection\n\nCircuit-breaker and outlier-detection payloads moved verbatim to [references\u002Fworked-examples.md](references\u002Fworked-examples.md).\nLoad on demand when configuring backendDefaults health checks or outlierDetection.\n\n## Step 6 — Virtual services\n\nA virtual service is the client-facing abstraction. It is backed by\neither a virtual node (direct, no splitting) or a virtual router\n(weighted routing).\n\n**Virtual service backed by a virtual router (canary-capable):**\n\n```bash\naws appmesh create-virtual-service \\\n  --mesh-name production-mesh \\\n  --virtual-service-name checkout.mesh.local \\\n  --spec '{\n    \"provider\": {\n      \"virtualRouter\": {\n        \"virtualRouterName\": \"checkout-router\"\n      }\n    }\n  }' \\\n  --region us-east-1\n```\n\n**Virtual service backed by a virtual node (no splitting):**\n\n```bash\naws appmesh create-virtual-service \\\n  --mesh-name production-mesh \\\n  --virtual-service-name inventory.mesh.local \\\n  --spec '{\n    \"provider\": {\n      \"virtualNode\": {\n        \"virtualNodeName\": \"inventory-v1\"\n      }\n    }\n  }' \\\n  --region us-east-1\n```\n\n**Critical:** a virtual service backed by a virtual node CANNOT do\nweighted routing. Only a virtual service backed by a virtual router\ncan split traffic. Choose the provider based on whether canary\u002Fblue-\ngreen is needed.\n\n## Step 7 — Virtual gateway (north-south ingress)\n\ncreate-virtual-gateway and create-gateway-route CLI moved verbatim to [references\u002Fworked-examples.md](references\u002Fworked-examples.md).\nLoad on demand when provisioning north-south ingress.\n\n## Step 8 — Envoy sidecar injection\n\nEKS webhook install, ECS sidecar JSON, and the critical no-Envoy failure mode moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nLoad on demand when injecting Envoy on any platform.\n\n## Step 9 — mTLS via ACM Private CA\n\nListener TLS and backend peer TLS (SDS) payloads moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nLoad on demand when enabling STRICT or PERMISSIVE mTLS.\n\n## Step 10 — Observability (CloudWatch + X-Ray)\n\nAccess-log payload and Envoy metric names moved verbatim to [references\u002Fworked-examples.md](references\u002Fworked-examples.md).\nLoad on demand when wiring observability.\n\n## Step 11 — Mesh scope (namespace vs cluster)\n\nNamespace-scoping kubectl commands moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nLoad on demand when scoping injection on EKS.\n\n## Step 12 — xDS protocol\n\nxDS flow walkthrough moved verbatim to [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md).\nKey implication kept here: route weight changes need no pod restarts — xDS pushes within seconds.\n\n## Step 13 — Recent features\n\nRecent AWS features (2023-2026) moved verbatim to [references\u002Fadvanced-patterns.md](references\u002Fadvanced-patterns.md).\nLoad on demand when evaluating newest App Mesh capabilities.\n\n## NEVER do these things\n\n1. **NEVER back a virtual service with a virtual node if you need\n   canary routing.** Only a virtual router can do weighted traffic\n   splitting. A virtual service backed by a virtual node sends ALL\n   traffic to that single node.\n\n2. **NEVER assume Envoy is auto-injected on all platforms.** Auto-\n   inject works ONLY on EKS with the App Mesh Controller and labeled\n   namespaces. ECS and EC2 require manual sidecar configuration.\n\n3. **NEVER use mTLS STRICT mode during migration.** STRICT rejects\n   all connections without valid certificates. Start with PERMISSIVE\n   (allows mixing), verify all services have certs, then switch to\n   STRICT.\n\n4. **NEVER forget the Cloud Map namespace and service.** Cloud Map\n   service discovery requires a pre-existing namespace and service.\n   Without them, the virtual node has no backends and returns 503.\n\n5. **NEVER use DROP_ALL egress filter without planning.** DROP_ALL\n   blocks all non-mesh egress. If services depend on databases,\n   external APIs, or S3, these connections will break unless explicit\n   virtual services are created for allowed egress.\n\n6. **NEVER assume route weight changes require pod restarts.** The\n   xDS protocol pushes configuration changes in near-real-time via\n   streaming. Weight changes take effect within seconds without\n   restarts.\n\n7. **NEVER confuse virtual node service discovery types.** DNS and\n   Cloud Map are mutually exclusive — a virtual node uses ONE or the\n   other. Mixing them is not supported.\n\n8. **NEVER forget that Envoy resolves the DNS hostname at runtime.**\n   DNS service discovery relies on the hostname being resolvable by\n   Envoy. If DNS is misconfigured, Envoy has no backends.\n\n9. **NEVER create a virtual gateway without gateway routes.** The\n   virtual gateway is the listener; gateway routes direct traffic to\n   virtual services. Without routes, no external traffic enters the\n   mesh.\n\n10. **NEVER set circuit breaker connection pool limits too low.**\n    Low limits eject healthy endpoints under normal load. Start with\n    generous limits and tune based on observed behavior.\n\n## Output format\n\n```text\nAPP_MESH: \u003Cmesh-name> — virtual service \u003Cvs-name> (\u003Cprotocol>)\nVERDICT: READY_TO_DEPLOY | PREREQUISITES_MISSING\nCHECKLIST:\n  [✓|✗] Service mesh: \u003Cmesh-name> (egress: ALLOW_ALL | DROP_ALL)\n  [✓|✗] Virtual node: \u003Cvn-name> (service discovery: DNS \u003Chostname> | Cloud Map \u003Cnamespace>\u002F\u003Cservice>)\n  [✓|✗] Virtual router: \u003Cvr-name> (listener: \u003Cport>\u002F\u003Cprotocol>)\n  [✓|✗] Weighted route: \u003Croute-name> — \u003Cvn-1> (\u003Cweight>%), \u003Cvn-2> (\u003Cweight>%)\n  [✓|✗] Virtual service: \u003Cvs-name> → backed by \u003Cvirtual-router | virtual-node>\n  [✓|✗] Retry policy: maxRetries=\u003Cn>, perRetry=\u003Cms>, events=\u003Cevent-list>\n  [✓|✗] Timeout: request=\u003Cvalue>\u003Cunit>\n  [✓|✗] Circuit breaker: healthCheck path=\u003Cpath>, outlier detection maxServerErrors=\u003Cn>\n  [✓|✗] Virtual gateway: \u003Cvgw-name> (listener: \u003Cport>\u002F\u003Cprotocol>)\n  [✓|✗] Gateway route: \u003Cgr-name> → \u003Cvirtual-service-name>\n  [✓|✗] Envoy sidecar: auto-inject (EKS namespace=\u003Cns>) | manual (ECS task | EC2)\n  [✓|✗] mTLS: STRICT | PERMISSIVE | disabled (ACM Private CA \u003Cca-arn>)\n  [✓|✗] Observability: CloudWatch metrics enabled, X-Ray tracing enabled\n  [✓|✗] Mesh ARN: arn:aws:appmesh:\u003Cregion>:\u003Cacct>:mesh\u002F\u003Cmesh-name>\nVERIFICATION_COMMANDS:\n  aws appmesh describe-mesh --mesh-name \u003Cmesh-name> --region \u003Cregion>\n  aws appmesh describe-virtual-node --mesh-name \u003Cmesh-name> --virtual-node-name \u003Cvn-name> --region \u003Cregion>\n  aws appmesh describe-route --mesh-name \u003Cmesh-name> --virtual-router-name \u003Cvr-name> --route-name \u003Croute-name> --region \u003Cregion>\n  aws appmesh describe-virtual-service --mesh-name \u003Cmesh-name> --virtual-service-name \u003Cvs-name> --region \u003Cregion>\n```\n\n### Worked example — weighted canary with Cloud Map discovery and Envoy auto-inject\n\n```text\nAPP_MESH: production-mesh — virtual service checkout.mesh.local (http)\nVERDICT: READY_TO_DEPLOY\nCHECKLIST:\n  [✓] Service mesh: production-mesh (egress: ALLOW_ALL)\n  [✓] Virtual node: checkout-v1 (service discovery: Cloud Map mesh-services\u002Fcheckout)\n  [✓] Virtual node: checkout-v2 (service discovery: Cloud Map mesh-services\u002Fcheckout-v2)\n  [✓] Virtual router: checkout-router (listener: 8080\u002Fhttp)\n  [✓] Weighted route: checkout-canary — checkout-v1 (90%), checkout-v2 (10%)\n  [✓] Virtual service: checkout.mesh.local → backed by virtual-router checkout-router\n  [✓] Retry policy: maxRetries=3, perRetry=2000ms, events=server-error,gateway-error\n  [✓] Timeout: request=15s\n  [✓] Circuit breaker: healthCheck path=\u002Fhealth, outlier detection maxServerErrors=5\n  [✓] Envoy sidecar: auto-inject (EKS namespace=default, webhook=appmesh-controller)\n  [✓] mTLS: PERMISSIVE (migration mode, ACM Private CA active)\n  [✓] Observability: CloudWatch metrics enabled, X-Ray tracing enabled\n  [✓] Mesh ARN: arn:aws:appmesh:us-east-1:123456789012:mesh\u002Fproduction-mesh\nVERIFICATION_COMMANDS:\n  aws appmesh describe-mesh --mesh-name production-mesh --region us-east-1\n  aws appmesh describe-virtual-node --mesh-name production-mesh --virtual-node-name checkout-v1 --region us-east-1\n  aws appmesh describe-route --mesh-name production-mesh --virtual-router-name checkout-router --route-name checkout-canary --region us-east-1\n  aws appmesh describe-virtual-service --mesh-name production-mesh --virtual-service-name checkout.mesh.local --region us-east-1\n```\n\n## Error handling\n\nError-handling deep dives (503s, canary not splitting, Envoy not injected, mTLS rejected, DROP_ALL breakage, empty Cloud Map) moved verbatim to [references\u002Ferror-handling.md](references\u002Ferror-handling.md).\nLoad on demand when a deployment or runtime failure must be diagnosed.\n\n## References (load on demand)\n\n- [references\u002Fservice-discovery-and-routing.md](references\u002Fservice-discovery-and-routing.md) — DNS vs Cloud Map and routing deep dive; now also holds Steps 2-4 (virtual nodes, weighted routes, route policies) moved from SKILL.md.\n- [references\u002Fenvoy-and-mtls-guide.md](references\u002Fenvoy-and-mtls-guide.md) — Envoy injection and mTLS deep dive; now also holds the Envoy auto-inject heuristic and Steps 8, 9, 11, 12 moved from SKILL.md.\n- [references\u002Fworked-examples.md](references\u002Fworked-examples.md) — secondary worked examples moved from SKILL.md: circuit breaker (Step 5), virtual gateway (Step 7), observability (Step 10).\n- [references\u002Ferror-handling.md](references\u002Ferror-handling.md) — error-handling deep dives moved from SKILL.md: 503 no healthy upstream, canary not splitting, Envoy not injected, mTLS rejections, DROP_ALL egress breakage, Cloud Map with no instances.\n- [references\u002Fadvanced-patterns.md](references\u002Fadvanced-patterns.md) — cross-dependency gotchas and Recent AWS features 2023-2026 (Step 13) moved from SKILL.md.\n\n## Domain\n\nAWS CloudOps \u002F AWS App Mesh Virtual Service Layer Provisioning &\nService Mesh Networking.\n\n## AWS documentation\n\n- **App Mesh User Guide** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fwelcome.html\n- **Virtual nodes** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fvirtual_nodes.html\n- **Virtual routers and routes** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fvirtual_routers.html\n- **Virtual services** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fvirtual_services.html\n- **Virtual gateways** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fvirtual_gateways.html\n- **Envoy proxy** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fenvoy.html\n- **mTLS** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fmtls.html\n- **Cloud Map service discovery** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fcloudmap.html\n- **App Mesh Controller for EKS** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fgetting-started-kubernetes.html\n- **CloudWatch metrics** — https:\u002F\u002Fdocs.aws.amazon.com\u002Fapp-mesh\u002Flatest\u002Fuserguide\u002Fmetrics.html\n",{"name":5,"license":15,"metadata":24,"description":7,"compatibility":41},{"tags":25,"phase":26,"author":27,"domain":28,"family":29,"version":30,"keywords":31,"task_type":32,"complexity":33,"entry_point":34,"skill_class":35,"when_to_use":36,"dependencies":37,"requires_llm":38,"verdict_shape":39,"lifecycle_status":40,"supports_pipeline":38},"aws, appmesh, service-mesh, virtual-service, envoy, networking, deploy, cloudops, weighted-routing, canary, cloud-map, mtls, circuit-breaking","1","Jacky Chan — AWS Community Builder","aws-cloudops","Networking","0.1.0","aws, app mesh, service mesh, virtual node, virtual service, virtual router, virtual gateway, weighted routing, canary deploy, traffic shifting, envoy proxy, envoy sidecar, auto inject, mutating webhook, xds protocol, cloud map, dns service discovery, service discovery, circuit breaker, outlier detection, connection pool, retry policy, timeout policy, mTLS, acm private ca, x-ray tracing, cloudwatch metrics, cloudops, deploy, networking","deploy","high","false","capability","Invoke when the user wants to create an App Mesh service mesh with virtual nodes, virtual services, and virtual routers, configure weighted routing for canary traffic shifting, set up a virtual gateway for north-south ingress, inject Envoy sidecars (auto-inject on EKS via webhook, ECS sidecar, or EC2 manual), enable mTLS via ACM Private CA, configure circuit breakers and outlier detection, or integrate CloudWatch metrics and X-Ray tracing. Do NOT invoke for AWS Cloud Map service discovery without App Mesh (use servicediscovery skills), Istio service mesh (different control plane), or AWS Gateway Load Balancer (different Layer 3\u002F4 inspection service).","aws-orchestrator","true","READY_TO_DEPLOY | PREREQUISITES_MISSING","active","Agent runtime that reads SKILL.md (Claude Code, Cursor, Windsurf, Codex, Gemini). For live deployment: AWS CLI v2 with appmesh create-mesh, create-virtual-node, create-virtual-router, create-route, create-virtual-service, create-virtual-gateway, create-gateway-route (AWS CLI v2, SSO or key-based credentials). For EKS Envoy injection: kubectl and the App Mesh Controller with mutating webhook configuration. For Cloud Map: servicediscovery create-service, create-http-namespace. For mTLS:...","main","eeace51bd5dd1db2030729d83eeab988861553aa","skills\u002Fappmesh-virtual-service-deployer",0,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills","https:\u002F\u002Fgithub.com\u002Fjhkchan","CTO & Co-Founder @Beever-AI  & @votee. CAISP. AWS Community Builder. Organizer@GDGCloudHK. MongoDB Champion. Technopreneur + Infornographer.","2026-09-24 08:54:10",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add jhkchan\u002Faws-agent-skills --skill appmesh-virtual-service-deployer",{"tool":56,"command":57},"bunx","bunx skills add jhkchan\u002Faws-agent-skills --skill appmesh-virtual-service-deployer",{"tool":59,"command":60},"pnpm","pnpm dlx skills add jhkchan\u002Faws-agent-skills --skill appmesh-virtual-service-deployer","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002FSKILL.md",true,[64,70,74,78,82,86,90,94,98,102,106,110,114,118,122,126,130,134,138,142],{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"eval\u002Ftest-cases.yaml","REFERENCE","参考文档",6155,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Feval\u002Ftest-cases.yaml",{"path":71,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":72,"sourceHtmlUrl":73},"evals\u002Fbaselines\u002Fcloud-map-service-discovery.md",760,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fbaselines\u002Fcloud-map-service-discovery.md",{"path":75,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":76,"sourceHtmlUrl":77},"evals\u002Fbaselines\u002Fenvoy-auto-inject-webhook.md",897,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fbaselines\u002Fenvoy-auto-inject-webhook.md",{"path":79,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":80,"sourceHtmlUrl":81},"evals\u002Fbaselines\u002Fmissing-cloudmap-namespace.md",778,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fbaselines\u002Fmissing-cloudmap-namespace.md",{"path":83,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":84,"sourceHtmlUrl":85},"evals\u002Fbaselines\u002Fvirtual-gateway-mtls.md",856,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fbaselines\u002Fvirtual-gateway-mtls.md",{"path":87,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":88,"sourceHtmlUrl":89},"evals\u002Fbaselines\u002Fweighted-canary-dns-discovery.md",1175,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fbaselines\u002Fweighted-canary-dns-discovery.md",{"path":91,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":92,"sourceHtmlUrl":93},"evals\u002Feval_config.yaml",1816,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Feval_config.yaml",{"path":95,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":96,"sourceHtmlUrl":97},"evals\u002Fevals.json",10968,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fevals.json",{"path":99,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":100,"sourceHtmlUrl":101},"evals\u002Fprompts\u002Fcloud-map-service-discovery.md",552,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fprompts\u002Fcloud-map-service-discovery.md",{"path":103,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":104,"sourceHtmlUrl":105},"evals\u002Fprompts\u002Fenvoy-auto-inject-webhook.md",573,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fprompts\u002Fenvoy-auto-inject-webhook.md",{"path":107,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":108,"sourceHtmlUrl":109},"evals\u002Fprompts\u002Fmissing-cloudmap-namespace.md",464,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fprompts\u002Fmissing-cloudmap-namespace.md",{"path":111,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":112,"sourceHtmlUrl":113},"evals\u002Fprompts\u002Fvirtual-gateway-mtls.md",535,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fprompts\u002Fvirtual-gateway-mtls.md",{"path":115,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":116,"sourceHtmlUrl":117},"evals\u002Fprompts\u002Fweighted-canary-dns-discovery.md",737,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fevals\u002Fprompts\u002Fweighted-canary-dns-discovery.md",{"path":119,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":120,"sourceHtmlUrl":121},"examples\u002FREADME.md",9184,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fexamples\u002FREADME.md",{"path":123,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":124,"sourceHtmlUrl":125},"references\u002Fadvanced-patterns.md",2284,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Freferences\u002Fadvanced-patterns.md",{"path":127,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":128,"sourceHtmlUrl":129},"references\u002Fenvoy-and-mtls-guide.md",18910,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Freferences\u002Fenvoy-and-mtls-guide.md",{"path":131,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":132,"sourceHtmlUrl":133},"references\u002Ferror-handling.md",1718,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Freferences\u002Ferror-handling.md",{"path":135,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":136,"sourceHtmlUrl":137},"references\u002Fservice-discovery-and-routing.md",14293,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Freferences\u002Fservice-discovery-and-routing.md",{"path":139,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":140,"sourceHtmlUrl":141},"references\u002Fworked-examples.md",3386,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Freferences\u002Fworked-examples.md",{"path":143,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":144,"sourceHtmlUrl":145},"skill.usf.yaml",4669,"https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fappmesh-virtual-service-deployer\u002Fskill.usf.yaml",[147,154,161,168,175,182],{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":153,"publishedAt":19},"01M7ADVD60X4PCAPKSE58P9NDC","xray-tracing-deployer","Deploys AWS X-Ray distributed tracing with production-grade configuration: X-Ray daemon deployment (EC2 systemd, ECS sidecar, EKS DaemonSet, Lambda built-in), SDK instrumentation (Java, Python, Node.js, Go, .NET) with annotations and metadata, sampling rules (default reservoir + rate vs custom rules with URL \u002F method \u002F service-name predicates), service map generation, trace query and retrieval, groups and insights, CloudWatch ServiceLens integration, IAM permissions (PutTraceSegments, GetSamplingRules), and latest features (Lambda Powertools tracing, OpenTelemetry \u002F ADOT Collector support, W3C trace context). Emits a READY_TO_DEPLOY checklist with every configuration item verified. Use when enabling X-Ray on a new application, adding tracing to ECS \u002F EKS \u002F Lambda, validating a sampling configuration, or generating daemon deployment CLI and IaC templates. Triggers: X-Ray, distributed tracing, X-Ray daemon, sampling rules, service map, X-Ray SDK, OpenTelemetry, ADOT, Lambda Powertools, CloudWatch ServiceLens.","https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fxray-tracing-deployer\u002FSKILL.md",[],[],{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":160,"publishedAt":19},"01M7ADTVJ7DCBM9CB1FY51GH7J","wellarchitected-workload-auditor","Audits AWS Well-Architected Tool workloads for review staleness, per-pillar high-risk issue counts, milestone tracking gaps, and remediation plan completeness. Emits a deterministic verdict (STALE_REVIEW | HIGH_RISK | CONFIG_GAP | OK) per workload with per-pillar risk breakdown, milestone trend analysis, and specific remediation. Use when reviewing Well-Architected workloads, checking WA review freshness, auditing high-risk issues per pillar, validating milestone coverage, or tracking remediation plans.","https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwellarchitected-workload-auditor\u002FSKILL.md",[],[],{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":167,"publishedAt":19},"01M7ADT59G6FTTNZYHT2QRPNNP","wellarchitected-review-operator","Operates AWS Well-Architected Tool reviews end-to-end — creates workloads, runs pillar reviews (operational excellence, security, reliability, performance, cost optimization, sustainability, Prosperity), answers questions with risk-tier rationale, generates reports and improvement plans, creates milestones, and integrates with Trusted Advisor and Well-Architected Labs. Runs deterministic pre-checks (workload existence, lens availability, pillar coverage), emits the exact update-answer \u002F get-consolidated-report CLI behind a CONFIRM gate, and verifies risk-tier transitions. Emits READY | BLOCKED | COMPLETED. Use when creating a workload, answering pillar questions, generating an improvement plan, creating a milestone, or integrating with Trusted Advisor. Triggers: well-architected tool, well architected review, workload review, pillar, operational excellence, security, reliability, performance, cost optimization, sustainability, prosperity, improvement plan, milestone, trusted advisor.","https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwellarchitected-review-operator\u002FSKILL.md",[],[],{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":172,"categories":173,"keywords":174,"publishedAt":19},"01M7ADSH5JA9QHNPPHQCD7EVA7","wafv2-web-acl-deployer","Provisions AWS WAFv2 Web ACLs with correct production defaults: scope selection (CLOUDFRONT vs REGIONAL), managed rule groups (AWSManagedRulesCommonRuleSet, SQLiRuleSet, LinuxRuleSet, WindowsRuleSet, IPReputationList, BotControlRuleSet, AmazonIpReputationList, ATP), custom rules (byte match, IP set, regex pattern, geo match, size constraint, rate-based), rule priority ordering, logging destinations (Kinesis Firehose, CloudWatch Logs, S3), resource association (ALB, API Gateway, CloudFront), rate-based rules with aggregate keys (IP, forwarded IP), Challenge and CAPTCHA actions, and account takeover prevention. Emits a READY_TO_DEPLOY checklist. Use when creating a Web ACL, attaching managed rules, configuring rate limiting, enabling WAF logging, associating a Web ACL with an ALB\u002FAPI Gateway\u002FCloudFront, or configuring CAPTCHA\u002FChallenge actions. Triggers: create WAF Web ACL, WAFv2, managed rule groups, WAF rate limiting, WAF logging, WAF CAPTCHA, WAF Challenge action, bot control, ATP managed rules.","https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwafv2-web-acl-deployer\u002FSKILL.md",[],[],{"skillNo":176,"name":177,"nameZh":6,"description":178,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":179,"categories":180,"keywords":181,"publishedAt":19},"01M7ADRYFAT8R73FBNCHRC573X","wafv2-web-acl-auditor","Audits AWS WAFv2 Web ACL configurations to determine whether the ACL provides effective protection: default-action posture (Allow vs Block), managed-rule-group coverage gaps, rule effectiveness (BLOCK vs COUNT, shadow\u002Fbypass rules, stale exclusions), rate-based rule correctness, logging and visibility configuration, and text-transformation bypass vectors. Emits a deterministic verdict per Web ACL. Use when reviewing a WAFv2 Web ACL for security posture, checking managed-rule coverage, validating rate-limit configuration, auditing WAF logging, or hardening a WAF before production deployment. Triggers: WAF, WAFv2, Web ACL, managed rule group, AWSManagedRules, rate-based rule, CAPTCHA, Challenge, OverrideAction, Count mode, forwarded IP, Bot Control, ATP, text transformation, WAF logging, WAF visibility, OWASP, SQLi, XSS, RFI.","https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwafv2-web-acl-auditor\u002FSKILL.md",[],[],{"skillNo":183,"name":184,"nameZh":6,"description":185,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":186,"categories":187,"keywords":188,"publishedAt":19},"01M7ADR8W7G72JX4KFGN03QK8Q","waf-rule-deployer","Provisions AWS WAFv2 rule sets with production defaults: Web ACL creation (CloudFront scope in us-east-1 vs Regional scope), rule groups (managed vs custom) with capacity planning in WCU (Web ACL Capacity Units, 1500 default per ACL), managed rule groups (AWSManagedRulesCommonRuleSet, AWSManagedRulesSQLiRuleSet, AWSManagedRulesAmazonIpReputationList, AWSManagedRulesLinuxRuleSet, AWSManagedRulesWindowsRuleSet, AWSManagedRulesUnixRuleSet), custom rules (byte-match, regex pattern set, geo-match, rate-based), IP set and regex pattern set resources, rule priority ordering (lowest number evaluated first), action types (allow, block, count, CAPTCHA, Challenge), bot control managed rule group, account takeover. Triggers: create WAF rule, WAF managed rule group, WAF custom rule, WAF byte match, WAF geo match, WAF rate based rule, WAF IP set, WAF regex pattern set, WAF WCU budget, WAF CloudFront ACL, WAF logging Kinesis Firehose, WAF bot control, WAF ATP, WAF label matching, WAF Challenge action, WAF CAPTCHA.","https:\u002F\u002Fgithub.com\u002Fjhkchan\u002Faws-agent-skills\u002Fblob\u002Fmain\u002Fskills\u002Fwaf-rule-deployer\u002FSKILL.md",[],[]]