Error Debugging
Security scan passedTwo debugging subagents and three commands that find the root cause of errors, set up error tracking and run multi-agent code reviews.

Build AI agents on Amazon Bedrock AgentCore with AWS's skills for scaffolding, tools, memory, evals, debugging and going live.
by Amazon· AI & agentsUnclaimed · Yours? Claim it
4 images
About
AWS Agents is Amazon Web Services' plugin for building AI agents on AWS, part of the Agent Toolkit for AWS. Its skills guide your coding agent through the whole life of an agent on Amazon Bedrock AgentCore, from a new project to production, and it connects the AWS Knowledge MCP Server for current AWS documentation. What its skills cover - Getting started: choosing a framework (Strands, LangGraph, Google ADK or OpenAI Agents), creating the project with the AgentCore CLI, then the first deploy and call - Building: memory across sessions, calling the agent from your own app, VPCs, multi-agent and A2A setups, browser and code interpreter tools, and moving from Bedrock Agents - Connecting tools through AgentCore Gateway (APIs, Lambda functions, OpenAPI and MCP servers), with outbound auth and Cedar policies that limit what the agent may do - Deploying: pre-flight checks, CDK, IAM and quota errors, versions, rollback and canary deploys - Debugging wrong answers, errors and timeouts from traces and logs, and checking your setup when the CLI fails - Measuring quality with evaluators and CI/CD quality gates, plus observability and cost - Hardening for production: inbound auth (JWT, SigV4), IAM scoping, secrets, rate limits and cold starts - Payments: letting an agent pay for x402-protected content through AgentCore Payments, within spend limits you set How to install it In Claude Code, run /plugin install aws-agents@claude-plugins-official, or add the aws/agent-toolkit-for-aws marketplace and install aws-agents@agent-toolkit-for-aws. It's also in Codex and Cursor through the aws/agent-toolkit-for-aws marketplace, and in Devin with devin plugins install aws/agent-toolkit-for-aws#plugins/aws-agents. What you need - The AgentCore CLI 0.9.0 or later (npm install -g @aws/agentcore), with Node.js 20 or later - The AWS CLI, with your credentials set up - Python 3.11 or later, with uv Good to know - The bundled AWS Knowledge MCP Server is a remote server for AWS documentation, the Strands Agents docs, regional availability and AWS skills. It needs no sign-in or AWS account, and it's rate-limited. - The agents you build are deployed to and run in your own AWS account.
Once it’s added
“How do I build an agent on AWS?”
“My agent keeps forgetting what I told it.”
“Deploy is failing with a CDK error.”
“I want to call my deployed agent from my React app.”
“Restrict my agent from making purchases over $1000.”
“We're going live next week. What should I check?”
Before you add it
Reads only
Current AWS and Strands Agents documentation, through the AWS Knowledge MCP Server · Your agent's traces and logs, when debugging · Your agent project and its setup
Makes changes · your agent can ask first
Creates and changes your agent project's code and configuration · Deploys, updates and rolls back agents in your AWS account with the AgentCore CLI
From people who use it
Step by step
Nothing new reaches your agent until a person has looked at it.
No reviews yet. Used AWS Agents? Say how it went.
Write a reviewRuns on your machine.
Runs on your machine.
/plugin marketplace add aws/agent-toolkit-for-aws /plugin install aws-agents@agent-toolkit-for-aws
Scanned and passed · checked 7 h ago
Plugins people install alongside AWS Agents.
Scanned on every change, and again when our rules improve.
Tell us and a person on the trust team will look at it. Or email [email protected].
Nobody has shared how they use AWS Agents yet.
Security scan
Passed
Safe to install in every agent it supports. Claude plugins.
Plugins people install alongside AWS Agents.
Two debugging subagents and three commands that find the root cause of errors, set up error tracking and run multi-agent code reviews.
ByteDance's local MCP server that lets your agent run any shell command or script on your machine, with your own permissions.
Keeps your agent's plan, findings and progress in markdown files on disk, so long tasks survive /clear, crashes and compaction.