20 SageMaker HyperPod Cluster Debugger alternatives
SageMaker HyperPod Cluster Debugger: Read-only diagnostics for cluster-wide SageMaker HyperPod failures on EKS or Slurm — CloudFormation errors, EFA health checks, lifecycle scripts, capacity, dangling nodes and autoscaler conflicts. 20 agent skills do similar work, most sharing Kubernetes, Official and GPU, ranked by how much they have in common with SageMaker HyperPod Cluster Debugger.
Back to SageMaker HyperPod Cluster Debugger
-
by NVIDIA
Deploy, configure and troubleshoot the NVIDIA RAG Blueprint from your agent — Docker Compose, Helm or library, with feature toggles for VLM, guardrails and agentic RAG.
Also: Kubernetes, Troubleshooting
-
by AWS Labs
Build multi-step Lambda workflows that survive interruption and run for up to a year — the replay model, step operations, wait/callback patterns, saga-style compensation and local testing.
-
by .NET Team at Microsoft
Diagnose an MSBuild build failure from a .binlog file — query the binary log through a dedicated MCP server instead of grepping console output, with a text-replay fallback.
Also: Official, Troubleshooting
-
by Microsoft
Microsoft's official AKS planning skill: separates the Day-0 decisions you cannot undo (networking, API server) from the Day-1 features you can switch on later.
Also: Kubernetes, Official
-
by AMD
AMD's official skill for standing up a vLLM endpoint on Instinct MI300X/MI325X/MI350X/MI355X GPUs — detection, recipe lookup, launch and health check in one flow.
-
by AMD
AMD's official skill for benchmarking vLLM/SGLang inference and driving GPU kernel optimization with Magpie and TraceLens trace analysis.
-
by Railway
Railway's official agent skill: create projects, provision databases and buckets, deploy, manage variables and domains, and read build failures back — from the CLI, API or MCP server.
Also: Infrastructure
-
by Google Cloud
Google's official skill for driving the gcloud CLI safely from an agent: validate every command against its own help text, cap the output, and refuse the operations that should never run unattended.
Also: Infrastructure
-
by Amazon Web Services
Official AWS skill for authoring, deploying and debugging CDK stacks — construct patterns, bootstrap, drift, resource import and the CloudFormation errors that trap people.
Also: AWS
-
by HashiCorp
HashiCorp's official Packer skill for building Amazon Machine Images with the amazon-ebs builder, covering template structure, plugin pinning and the cost and timing caveats.
Also: AWS
-
by Qdrant
Qdrant's meta-skill: instead of shipping static docs, it loads the current official Qdrant skill tree live from skills.qdrant.tech and diagnoses from that.
Also: Troubleshooting
-
by Grafana Labs
Grafana's official Alloy skill: write one collector config that carries metrics, logs, traces and profiles, validate it locally, and prove samples are actually leaving the box.
Also: Kubernetes
-
by Grafana Labs
Grafana's official Tempo skill: stand up a tracing backend that needs nothing but object storage, then write TraceQL that finds the slow span instead of scrolling for it.
Also: Kubernetes
-
by Grafana Labs
Grafana's official Loki skill: write LogQL that returns rows instead of timeouts, ship logs through Alloy, and reason about a store that indexes labels rather than log text.
Also: Kubernetes
-
by NVIDIA
NVIDIA's official skill for bringing up Dynamo inference routing — pick a router mode, enable KV-aware routing, and smoke-test the frontend endpoint before claiming anything works.
Also: Kubernetes
-
by .NET Foundation
The .NET team's official skill for capturing crash dumps from CoreCLR and NativeAOT apps on Linux, macOS, Windows and inside containers — collection only, no analysis.
Also: Kubernetes
-
by Google
Google's official GKE skill — when Autopilot beats Standard, Workload Identity instead of JSON keys, Autopilot's 250m CPU rounding, and private-cluster flags.
Also: Kubernetes
-
by Datadog
Install, instrument and analyse Datadog APM from an agent — Kubernetes and Linux Single-Step Instrumentation, service remapping, trace search and service maps.
Also: Kubernetes
-
by NVIDIA
Inspects the host's hardware, OS, CUDA driver and existing tooling, recommends one Holoscan SDK install method with a reason, then hands off to the matching install skill.
Also: GPU
-
by NVIDIA
Picks the right vLLM or SGLang runtime for your Jetson generation and JetPack version, then produces a working OpenAI-compatible serving command.
Also: GPU
