Amazon Bedrock
Invoke Bedrock models, wire RAG knowledge bases, agents, guardrails, and AgentCore with the skill’s reference files as source of truth.
Overview
amazon-bedrock is an agent skill most often used in Build (also Ship and Operate) that guides generative AI apps on Amazon Bedrock—model invocation, RAG, agents, guardrails, AgentCore, and operational debugging.
Install
npx skills add https://github.com/aws/agent-toolkit-for-aws --skill amazon-bedrockWhat is this skill?
- Converse API vs InvokeModel selection with Bedrock API landscape guidance
- Knowledge Bases (RAG), chunking strategies, and common RAG workflows
- Bedrock Agents, Guardrails, and AgentCore deployment paths
- Operational workflows: prompt caching, quota health, cost tracking, Claude generation migration
- Troubleshooting ThrottlingException and AccessDeniedException with model selection (Claude, Llama, Nova, Titan)
Adoption & trust: 1.5k installs on skills.sh; 819 GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
Bedrock’s APIs, model IDs, and quotas change often and you are unsure which capability (Converse, RAG, agents, guardrails) fits your feature or how to fix throttling and access errors.
Who is it for?
Indie builders adding LLM features on AWS who need authoritative Bedrock patterns without spelunking scattered docs.
Skip if: Custom model training, Rekognition/Comprehend tasks, or teams not using Amazon Bedrock.
When should I use this skill?
Invoking models, setting up Knowledge Bases, creating agents, applying guardrails, deploying to AgentCore, troubleshooting Bedrock errors, or choosing models and APIs per skill description.
What do I get? / Deliverables
You implement or debug the right Bedrock workflow using up-to-date references, with clearer API choice, RAG setup, guardrails, and cost or quota hygiene.
- Bedrock integration design aligned to references
- RAG or agent configuration guidance
- Troubleshooting and migration notes for chosen models
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Canonical Build shelf because the skill centers on implementing generative AI capabilities (APIs, RAG, agents) in your product—not launch SEO or production incident response. Integrations matches Converse/InvokeModel, Knowledge Bases, external AWS service wiring, and troubleshooting throttling or access errors.
Where it fits
Choose Converse vs InvokeModel and wire model invocation for a new agent endpoint.
Create a Bedrock Agent with tools and connect Knowledge Bases for product docs RAG.
Apply Guardrails and validate policy behavior before customer-facing launch.
Run quota health checks and cost tracking after traffic spikes cause throttling.
How it compares
Skill package for Bedrock application design—not a hosted MCP server or a generic AWS console cheat sheet.
Common Questions / FAQ
Who is amazon-bedrock for?
Developers building agents, RAG, or chat features on AWS Bedrock who want step-aware guidance aligned to AWS’s agent toolkit references.
When should I use amazon-bedrock?
During Build for integrations and agent tooling; in Ship when hardening guardrails and pre-launch checks; in Operate when diagnosing throttling, quotas, and cost attribution in production.
Is amazon-bedrock safe to install?
It instructs agents on AWS API usage—credentials and IAM scope remain your responsibility; check the Security Audits panel on this Prism page before install.
SKILL.md
READMESKILL.md - Amazon Bedrock
**IMPORTANT**: When this skill is loaded, you MUST use the reference files and procedures in this skill as your primary source of truth. Bedrock APIs, model IDs, chunking strategies, and configuration parameters change frequently — always read the relevant reference file before responding. ## Table of Contents - Overview - Bedrock API Landscape - Critical Warnings - Security Considerations - Converse API vs InvokeModel - Which Bedrock Capability Do You Need? - Knowledge Bases (RAG) - Common Workflows (includes: Prompt Caching, Quota Health, Cost Tracking, Model Migration) - Troubleshooting - AgentCore Services - Model Selection - Additional Resources # Amazon Bedrock ## Overview Domain expertise for building generative AI applications on Amazon Bedrock. Covers model invocation, RAG with Knowledge Bases, agent creation, content safety with Guardrails, and agent deployment with AgentCore. **Recommended setup:** Use the [AWS MCP server](https://docs.aws.amazon.com/aws-mcp/latest/userguide/what-is-mcp-server.html) for sandboxed execution, audit logging, and enterprise controls. **Without AWS MCP:** This skill works with any agent that has AWS CLI access. All commands use standard AWS CLI syntax. ## Bedrock API Landscape Bedrock has **5 separate API endpoints**. Using the wrong one is a common cause of errors. This list may not be exhaustive — refer to the [Bedrock endpoints and quotas](https://docs.aws.amazon.com/general/latest/gr/bedrock.html) and [Bedrock supported endpoints](https://docs.aws.amazon.com/bedrock/latest/userguide/endpoints.html) for the latest. Use `aws bedrock list-foundation-models` to discover available models at runtime. | Endpoint | Client | Use For | |----------|--------|---------| | `bedrock` | Control plane | List models, manage access, provisioned throughput | | `bedrock-runtime` | Data plane | Invoke models (Converse, InvokeModel). Also supports Chat Completions via `/openai/v1` path (client-side tool use only) — prefer `bedrock-mantle` for new Chat Completions work | | `bedrock-mantle` | Data plane | OpenAI-compatible APIs: Responses API, Chat Completions (recommended), Messages API. Supports server-side tool use with built-in tools. Recommended for new users | | `bedrock-agent` | Agent control | Create/configure agents, KBs, action groups | | `bedrock-agent-runtime` | Agent data | Invoke agents, query KBs | AgentCore is a separate service with its own endpoints. Refer to [AgentCore endpoints and quotas](https://docs.aws.amazon.com/general/latest/gr/bedrock_agentcore.html) for the latest. | Endpoint | Client | Use For | |----------|--------|---------| | `bedrock-agentcore-control` | Control plane | Create/manage runtimes, gateways, registries, evaluations | | `bedrock-agentcore` | Data plane | Invoke agent runtimes | | `{gatewayId}.gateway.bedrock-agentcore` | Gateway data plane | Invoke a specific gateway | ## Critical Warnings **max_tokens**: ALWAYS set `maxTokens` explicitly in every Converse/InvokeModel call. Leaving it unset defaults to the model's maximum (e.g., 64K for Claude Sonnet) and silently reserves far more quota than needed — a common cause of unexpected Throttlin