amazon-bedrock

Guide building and operating generative AI applications on Amazon Bedrock.

Updated May 12, 2026
One-click install
npx skills add https://github.com/mreferre/aws-agent-toolkit-skills --skill amazon-bedrock-mreferre
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: amazon-bedrock
Source: https://github.com/mreferre/aws-agent-toolkit-skills/tree/main/.kiro/skills/amazon-bedrock
Command: npx skills add https://github.com/mreferre/aws-agent-toolkit-skills --skill amazon-bedrock-mreferre

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It helps you design, invoke, secure, and troubleshoot generative AI applications on Amazon Bedrock without getting stuck on changing APIs, quotas, or model-specific request formats.

Core Features & Use Cases

  • Model Invocation (Converse/InvokeModel): Choose the right Bedrock runtime API, set required parameters like maxTokens, and handle throttling correctly.
  • RAG with Knowledge Bases: Create Knowledge Bases, configure chunking/vector stores, ingest data, and query with retrieve modes and metadata filtering.
  • Agents, Guardrails, and AgentCore Deployment: Build Bedrock Agents with action groups, apply guardrails safely, and deploy and scale via AgentCore (Gateway/Runtime/Memory/Observability/Registry/Evaluations).

Quick Start

Invoke a Claude model via the Bedrock Converse API with an explicit maxTokens setting and a region-qualified model ID, then iterate using the same skill references for troubleshooting and optimization.

Frequently Asked Questions about amazon-bedrock

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I choose between Converse and InvokeModel APIs for Amazon Bedrock?

To choose between Converse and InvokeModel on Amazon Bedrock, evaluate whether you need a unified conversational interface across models or direct access to model-specific parameters and payload formats.

How do I set up a RAG Knowledge Base in Amazon Bedrock?

Setting up a RAG Knowledge Base in Amazon Bedrock involves creating the base, configuring chunking strategies and vector stores, ingesting data, and querying using retrieve modes with metadata filtering.

Does Amazon Bedrock require explicitly setting maxTokens for Claude model invocation?

Yes, Amazon Bedrock enforces required operational constraints by mandating explicit maxTokens settings when invoking Claude models to prevent runaway generation costs and unexpected truncation.

What is the best way to troubleshoot Bedrock quota and throttling issues?

Troubleshooting Amazon Bedrock quota and throttling issues requires performing quota health checks, validating correct endpoint and client usage, and following step-by-step procedures for deterministic configuration.

Can I deploy and scale Bedrock Agents using AgentCore?

Yes, you can deploy and scale Amazon Bedrock Agents via AgentCore by utilizing its Gateway, Runtime, Memory, Observability, Registry, and Evaluations components for production-grade operations.

How do I apply guardrails safely when building Amazon Bedrock Agents?

Applying guardrails safely to Amazon Bedrock Agents involves configuring action groups and attaching guardrail policies to filter harmful content and enforce responsible AI constraints during invocation.