sagemaker-ai plugin
Build, train, and deploy AI models with deep AWS AI/ML expertise brought directly into your coding assistants, covering the surface area of Amazon SageMaker AI.
How to install the sagemaker-ai plugin
Run both commands inside Claude Code. The first adds the marketplace once; the second installs the plugin.
/plugin marketplace add awslabs/agent-plugins
/plugin install sagemaker-ai@agent-plugins-for-awsWhat's inside sagemaker-ai
20 parts read from the plugin's own files.
Skills 19
Claude loads it when the task matches- dataset-evaluation
Validates dataset formatting and schema compliance for SageMaker model fine-tuning strategies like SFT, DPO, or RLVR.
- dataset-transformation
Generates Python code to transform datasets between various machine learning schemas for model training or evaluation.
- directory-management
Sets up project directory structures and organizes artifacts like specs, scripts, notebooks, and manifests.
- finetuning
Generates code to fine-tune base models using SageMaker serverless training jobs for various techniques like SFT, DPO, RLVR, and RLAIF.
- finetuning-technique
Selects and validates fine-tuning techniques like SFT, DPO, RLVR, or RLAIF based on a user's use case and model compatibility.
- hyperpod-cluster-debugger
Diagnose and remediate HyperPod cluster issues including CloudFormation failures, EFA health checks, EKS access, and node replacement.
- hyperpod-issue-report
Generates diagnostic issue reports from HyperPod EKS and Slurm clusters by collecting logs and configurations via SSM and storing them in S3.
- hyperpod-nccl
Diagnose NCCL failures and training-pod issues on HyperPod GPU clusters running EKS or Slurm.
- hyperpod-node-debugger
Diagnose and remediate per-node issues on HyperPod clusters including EKS or Slurm, covering GPU hardware, EFA, networking, and resource exhaustion.
- hyperpod-performance-debugger
Diagnoses performance issues on HyperPod clusters related to uneven NCCL bandwidth and poor filesystem throughput.
- hyperpod-slurm-debugger
Diagnoses Slurm scheduler and node-daemon issues on HyperPod Slurm clusters using read-only diagnostic scripts.
- hyperpod-ssm
Execute remote commands and transfer files on SageMaker HyperPod cluster nodes using AWS Systems Manager.
- hyperpod-version-checker
Check and compare software component versions across SageMaker HyperPod cluster nodes to verify compatibility and detect mismatches.
- model-deployment
Generates deployment code for fine-tuned Nova and OSS models to SageMaker endpoints or Amazon Bedrock.
- model-evaluation
Generates Python code to evaluate SageMaker models using LLM-as-Judge or Custom Scorer methods.
- model-selection
Selects and evaluates base models by querying SageMaker Hub to resolve model IDs, licenses, and benchmarks.
- planning
Discovers user intent and generates structured, step-by-step plans for model customization workflows like fine-tuning and training.
- sdk-getting-started
Validates the user's environment by checking the SDK version, AWS region, and execution role for SageMaker operations.
- use-case-specification
Generates a structured Markdown document defining business problems, stakeholders, and measurable success criteria for model customization.
MCP servers 1
Starts with the plugin- mcp
No description in the file.
Using sagemaker-ai in Claude Code
Skills need no command. Claude reads them when your request matches what they describe.
Update, turn off or remove
- Update the plugin (shell)
- claude plugin update sagemaker-ai@agent-plugins-for-aws
- Refresh the marketplace
- /plugin marketplace update agent-plugins-for-aws
- Turn it off for now
- /plugin disable sagemaker-ai@agent-plugins-for-aws
- Remove it
- /plugin uninstall sagemaker-ai@agent-plugins-for-aws
Skills in sagemaker-ai
Validates dataset formatting and schema compliance for SageMaker model fine-tuning strategies like SFT, DPO, or RLVR.
4 files938starsGenerates Python code to transform datasets between various machine learning schemas for model training or evaluation.
6 files938starsGenerates code to fine-tune base models using SageMaker serverless training jobs for various techniques like SFT, DPO, RLVR, and RLAIF.
12 files938starsDiagnose and remediate HyperPod cluster issues including CloudFormation failures, EFA health checks, EKS access, and node replacement.
8 files938starsGenerates diagnostic issue reports from HyperPod EKS and Slurm clusters by collecting logs and configurations via SSM and storing them in S3.
4 files938starsDiagnose NCCL failures and training-pod issues on HyperPod GPU clusters running EKS or Slurm.
6 files938starsDiagnose and remediate per-node issues on HyperPod clusters including EKS or Slurm, covering GPU hardware, EFA, networking, and resource exhaustion.
7 files938starsDiagnoses performance issues on HyperPod clusters related to uneven NCCL bandwidth and poor filesystem throughput.
3 files938starsDiagnoses Slurm scheduler and node-daemon issues on HyperPod Slurm clusters using read-only diagnostic scripts.
3 files938starsExecute remote commands and transfer files on SageMaker HyperPod cluster nodes using AWS Systems Manager.
5 files938starsCheck and compare software component versions across SageMaker HyperPod cluster nodes to verify compatibility and detect mismatches.
2 files938starsGenerates deployment code for fine-tuned Nova and OSS models to SageMaker endpoints or Amazon Bedrock.
11 files938starsGenerates Python code to evaluate SageMaker models using LLM-as-Judge or Custom Scorer methods.
15 files938starsSelects and evaluates base models by querying SageMaker Hub to resolve model IDs, licenses, and benchmarks.
4 files938starsDiscovers user intent and generates structured, step-by-step plans for model customization workflows like fine-tuning and training.
5 files938starsGenerates a structured Markdown document defining business problems, stakeholders, and measurable success criteria for model customization.
938starsSelects and validates fine-tuning techniques like SFT, DPO, RLVR, or RLAIF based on a user's use case and model compatibility.
3 files938starsValidates the user's environment by checking the SDK version, AWS region, and execution role for SageMaker operations.
3 files938starsSets up project directory structures and organizes artifacts like specs, scripts, notebooks, and manifests.
938stars
sagemaker-ai plugin questions
How do I install the sagemaker-ai plugin?
In Claude Code run /plugin marketplace add awslabs/agent-plugins, then /plugin install sagemaker-ai@agent-plugins-for-aws.
What does the sagemaker-ai plugin include?
sagemaker-ai bundles 19 skills, MCP servers.
How do I update or remove the sagemaker-ai plugin?
Update it with claude plugin update sagemaker-ai@agent-plugins-for-aws in your shell, or from the Installed tab of /plugin. Remove it with /plugin uninstall sagemaker-ai@agent-plugins-for-aws.