# AWS adds SageMaker inference‑optimizing skill for coding agents

> Amazon SageMaker now offers an aws‑ai‑ml skill that lets coding agents generate and benchmark deployment code for generative AI models.

Oossa · 2026-10-05 · https://oossa.com/en/aws-adds-sagemaker-inference-optimizing-skill-for-coding-agents

Amazon SageMaker launched a new skill called aws‑ai‑ml, available through the Agent Toolkit for AWS. The skill plugs into any coding agent that supports the Model Context Protocol, such as Kiro, Claude Code, or Codex, and lets the agent produce SageMaker Python SDK v3 code, benchmark endpoints, and recommend deployment configurations. Installation takes about 10 minutes on a local machine or in SageMaker Studio. The skill runs under the user’s AWS credentials, so no extra IAM setup is required.

## The facts

- Skill name: aws‑ai‑ml, announced Oct 5, 2026
- Works with agents that support MCP (Model Context Protocol)

## Why it matters

Developers can get ready‑to‑run SageMaker deployment code and performance data without manually tuning instance types, saving time and reducing guesswork.

## Sources & references

1. [New agent skill: Amazon SageMaker optimized generative AI inference for your coding agent](https://aws.amazon.com/blogs/machine-learning/new-agent-skill-amazon-sagemaker-optimized-generative-ai-inference-for-your-coding-agent/) – AWS, 2026-10-05

Last updated: 2026-10-05