Does this skill support streaming responses?

Yes, it provides implementation patterns for both native and Converse API streaming, allowing for real-time token delivery and better user experiences.

What is the benefit of using the Converse API?

The Converse API provides a unified, model-agnostic interface, allowing you to switch between different foundation models without rewriting your integration logic.

What is the Amazon Bedrock Inference skill?

It is a specialized capability that allows Claude Code to interact with the Amazon Bedrock Runtime API, enabling the invocation of foundation models like Claude, Nova, and Titan.

Are security guardrails included?

Yes, the skill covers the implementation of Bedrock Guardrails for runtime content filtering and safety compliance.

Which models are supported by this skill?

It supports a wide range of models available on Bedrock, including Anthropic Claude (Opus, Sonnet, Haiku), Amazon Nova, Amazon Titan, and third-party models from Meta and Cohere.

Amazon Bedrock Inference

Name: Amazon Bedrock Inference
Author: adaptationio

byadaptationio

クラウドインフラストラクチャ

Invokes foundation models like Claude, Nova, and Titan through the Amazon Bedrock Runtime API for advanced AI integrations.

概要

The Amazon Bedrock Inference skill provides a standardized framework for interacting with foundation models hosted on AWS. It supports both synchronous and asynchronous inference, streaming responses for low-latency user experiences, and the unified Converse API for model-agnostic implementations. This skill is essential for developers building production-grade conversational AI, implementing safety guardrails, and optimizing token usage across a range of high-performance models including Claude 3.5/4.5, Amazon Nova, and Titan.

主な機能

Cross-region inference profiles for high availability and cost optimization
0 GitHub stars
Built-in runtime guardrails for content safety and filtering
Real-time token streaming for low-latency user experiences
Long-running asynchronous inference support for batch processing
Unified Converse API for consistent multi-model interaction

ユースケース

Building production-grade conversational AI assistants
Implementing multi-turn dialogue with context-aware memory
Automating large-scale content generation with async workflows

概要

主な機能

Cross-region inference profiles for high availability and cost optimization
0 GitHub stars
Built-in runtime guardrails for content safety and filtering
Real-time token streaming for low-latency user experiences
Long-running asynchronous inference support for batch processing
Unified Converse API for consistent multi-model interaction

ユースケース

Building production-grade conversational AI assistants
Implementing multi-turn dialogue with context-aware memory
Automating large-scale content generation with async workflows