* feat(aws-startup-advisor): add claude-to-bedrock skill for moving Claude Desktop and Claude Code to Bedrock Add a claude-to-bedrock skill that moves the Claude apps a team uses, Claude Desktop (Chat and Cowork) and the Claude Code CLI, from Anthropic direct to Claude on Amazon Bedrock without losing claude.ai history. The skill asks first whether the person is an individual, an IT admin rolling out with Jamf or Intune, or a user on an IT-managed computer, checks the machine read-only, then walks only that route. History comes across through Claude Desktop's Import from Claude wizard, and into Claude Code through a converted export. Fleet scripts merge into an existing managed-settings.json instead of replacing it. The skill includes a repository-style description, a Last updated note, and a short section on what changes for the user (AWS billing and Activate credits, features not available on Bedrock). Also: - Routing: llm-to-bedrock hands Claude-app requests with no application code to claude-to-bedrock; architect-for-startups, start-building-for-startups, prompt-library-for-startups, and knowledge-base-for-startups route to it. Script paths resolve from <SKILL_BASE>. - README and all plugin manifests: list the skill and add keywords. - Telemetry: mark claude-to-bedrock NOT_YET_MODELLED, like bedrock-migration-assessment and render-to-aws, until PluginSkillId has a member. - Tests: port the export-script tests into tests/. * feat(aws-startup-advisor): add Definitions and a one-notice consent step to claude-to-bedrock Match the plugin's guided-skill conventions (operate-on-aws): a Definitions section (Load, <SKILL_BASE>, HUMAN STEP, routes), and a CONFIRM step after the read-only Phase 0 checks. It shows what will change, the AWS test call and billing, and how to undo it, then asks once and waits. It is adjusted per route, and a second ask is required for anything not listed (IAM resources, the Bedrock use case form). * feat(aws-startup-advisor): let claude-to-bedrock do the terminal part of each human step Sync from the upstream sample: - Add references/terminal-helpers.md: open pages and files for the person, quit Claude Desktop safely, check Bedrock model access read-only with get-use-case-for-model-access, and run aws sso login in the background. - aws-setup.md: check model access first, and add an optional, ask-first CLI path to create the Identity Center permission set. It runs a preflight, has the person retype the account ID, fails fast, and includes an undo block. - SKILL.md: point each step at the helpers. CONFIRM lists what they do, and says IAM or Identity Center changes always need their own yes. * feat(aws-startup-advisor): claude-to-bedrock uses sibling skills for model access and quota, and the shared model registry - Model access: when llm-to-bedrock is installed and uv is available, run its Access-only mode inline (preset model IDs and region). Its preflight probes each model and names the fix. Fall back to the read-only terminal helper when it isn't. - Quota: new Phase 4b for teams. Estimate TPM (about 20,000 per active developer as a starting point, plus output-token burndown from the vendored bedrock-quotas.md), check current quotas read-only with service-quotas, and request more only with its own yes. Uses prompt-library-for-startups' Bedrock Quota Manager when installed, otherwise shows the request-service-quota- increase command. - Models: pick Claude IDs from the vendored ai-model-lifecycle.md (Active rows only). The defaults are swapped everywhere if one goes Legacy. - Vendor ai/ai-model-lifecycle.md and ai/bedrock-quotas.md with a README table, so sync-vendored keeps them current. - llm-to-bedrock: fix the routing note. claude-to-bedrock calls its Access-only mode, not the other way around. - Sibling checks use <SKILL_BASE>/../<skill>/SKILL.md, so single-skill installs still work. * fix(aws-startup-advisor): claude-to-bedrock export script writes ASCII-only file names On Linux with a non-UTF-8 locale (LANG=C, common in containers) the OS can't encode file names like "café.md", so process_export.py crashed. Fold accents to their base letter and drop other non-ASCII characters in file names; the file contents stay UTF-8. Add a test that checks the names directly, since macOS accepts non-ASCII names and wouldn't fail on its own. Caught by CI. * fix(aws-startup-advisor): keep the plugin description unchanged and route Claude API code away from claude-to-bedrock Restore the Startup Advisor description in the manifests, marketplaces, and README. Add a first check in claude-to-bedrock that sends code calling the Claude API to llm-to-bedrock or bedrock-migration-assessment, and asks one question when the request is ambiguous. Drop the hand-off to the removed Bedrock Quota Manager prompt; Phase 4b uses its own CLI step. * refactor(aws-startup-advisor): rename claude-to-bedrock to claude-apps-to-bedrock The skill moves the Claude apps a team uses, not code that calls the Claude API. The new name keeps it apart from llm-to-bedrock at a glance. * fix(aws-startup-advisor): claude-apps-to-bedrock MDM profile scripts no longer duplicate an sso-session, and the feature notice matches Anthropic's 3P list The Jamf and Intune B2 scripts checked only the profile header, so a config that already had the session section got a second one and the AWS CLI rejected the whole file. Each section is now added only if its header is missing, under a skill-specific session name. The notice about features on Bedrock now uses the limits on Anthropic's Claude Desktop 3P feature page. * fix(aws-startup-advisor): address claude-apps-to-bedrock review findings - Claude Code imports a small HISTORY-INDEX.md (capped at 16 KB) instead of the whole history; GLOBAL-INSTRUCTIONS.txt is capped at 4 KB. - process_export.py writes 0700/0600 output and always deletes the extracted export; it tells the person to delete the zips when done. - Fleet profiles set deploymentMode=3p and disableDeploymentModeChooser so users can't leave the Bedrock configuration at sign-in. - Templates use the ClaudeBedrockInference permission set from 2b. - The 2b preflight collects names before grep, so it stops wherever the existing permission set appears; tests cover first, last, and absent. - Desktop steps follow Anthropic's docs: Enable Developer Mode, Apply Changes then Save & Restart, choose the 3P option at sign-in. - Phase 4b uses Anthropic's per-user TPM table, plans with 10x burndown for unlisted models, and notes that increases can be denied. - Desktop minimum is 4.23118.0; merge scripts keep a first-run .orig backup; the consent notice names the default model; bearer-token warning. * fix(aws-startup-advisor): keep @ in chat titles from becoming Claude Code imports HISTORY-INDEX.md is imported by ~/.claude/CLAUDE.md, and Claude Code follows @path imports inside it. Titles and summaries now use a full-width @ so a title like "Debug @~/.zshrc" doesn't load that file into every session. * chore(aws-startup-advisor): leave the root marketplace keywords unchanged The skill's keywords stay in the plugin's own manifests. The root marketplace listing is updated in release PRs. --------- Co-authored-by: Shrey Kapoor <kapurshr@amazon.com>
Agent Toolkit for AWS
Help AI coding agents build, deploy, and manage applications on AWS.
The Agent Toolkit for AWS gives AI coding agents the tools, knowledge, and guardrails they need to work with AWS services. It works with the coding agents developers already use — including Claude Code, Codex, Cursor, Devin, Kiro, and fx.
Quick start
AWS CLI
Use the Agent Toolkit directly from your terminal with the AWS CLI:
aws configure agent-toolkit
See the AWS CLI integration guide for setup, configuration, and usage instructions.
Claude Code
The plugins are available on the official Anthropic marketplace (claude-plugins-official) which is added to your Claude Code installation by default.
Use the following commands to install supported plugins from the toolkit:
For aws-core that covers service selection, CDK/CloudFormation, serverless, containers, storage, observability, billing, SDK usage, and deployment:
/plugin install aws-core@claude-plugins-official
Tip: If you get
Plugin not found, update your local marketplace index first:/plugin marketplace update claude-plugins-official
For aws-agents that covers building AI agents on AWS with Amazon Bedrock and AgentCore:
/plugin install aws-agents@claude-plugins-official
For aws-data-analytics that covers data lake, analytics, and ETL workflows with S3 Tables, AWS Glue, and Athena:
/plugin install aws-data-analytics@claude-plugins-official
For aws-agents-for-devsecops used to investigate incidents, review code and execute UAT for release readiness, scan code for vulnerabilities, and run penetration tests with AWS DevOps Agent and AWS Security Agent.
/plugin marketplace add aws/agent-toolkit-for-aws
/plugin install aws-agents-for-devsecops
/reload-plugins
# Or from Claude's official marketplace:
/plugin install aws-agents-for-devsecops@claude-plugins-official
/reload-plugins
# Setup:
/aws-agents-for-devsecops:setup
For aws-continuum that discovers, prioritizes, validates, and remediates security risks across the software development lifecycle with AWS Continuum:
/plugin marketplace add aws/agent-toolkit-for-aws
/plugin install aws-continuum@agent-toolkit-for-aws
/reload-plugins
# Setup:
/aws-continuum:setup
For aws-startup-advisor that covers startup-focused architecture, cost, security, and AWS Activate guidance, plus migrations to AWS from Azure, GCP, Heroku, and AI stacks such as OpenAI and Gemini, install from this repository's marketplace:
/plugin marketplace add aws/agent-toolkit-for-aws
/plugin install aws-startup-advisor@agent-toolkit-for-aws
/reload-plugins
Codex
In your terminal:
codex plugin marketplace add aws/agent-toolkit-for-aws
Then launch Codex and run /plugins to browse and install the aws-core plugin.
Cursor
Add this repository as a team marketplace from Settings → Plugins → Team Marketplaces → Add Marketplace → Import from Repo, pointing it at aws/agent-toolkit-for-aws. Cursor indexes the plugins listed in .cursor-plugin/marketplace.json on import.
Then open the Plugins panel and install the aws-core plugin (start here), or aws-agents and aws-data-analytics as needed. Each plugin bundles the AWS MCP Server configuration and agent skills.
Devin
In your terminal:
devin plugins install aws/agent-toolkit-for-aws#plugins/aws-core
This requires devin auth login and adds the plugin to your personal plugins, so it also loads in Devin cloud sessions and Devin Desktop. Add --local to install on this machine only. To install a different plugin, replace aws-core with its name from Plugins, such as aws-agents or aws-data-analytics.
In the Devin web app, choose Customize → Plugins → Add plugin → From repository, enter aws/agent-toolkit-for-aws, and set the subdirectory to plugins/aws-core (or another plugin's directory). Personal plugins added there are also available in the Devin CLI and Devin Desktop.
Kiro
Kiro setup has two independent parts: the AWS MCP Server (for runtime AWS API access and documentation search) and local skills (for task-specific agent guidance). They complement each other but work independently — skills don't require the MCP server, and the MCP server doesn't serve locally-installed skills.
1. Add the AWS MCP Server to your Kiro MCP configuration (.kiro/settings/mcp.json):
{
"mcpServers": {
"aws": {
"command": "uvx",
"args": [
"mcp-proxy-for-aws-cli@latest",
"https://aws-mcp.us-east-1.api.aws/mcp",
"--metadata",
"AWS_REGION=us-west-2"
]
}
}
}
The MCP server gives your agent access to AWS APIs, sandboxed script execution, and real-time documentation search.
2. Install skills from this repository:
npx skills add aws/agent-toolkit-for-aws/skills
This installs skill files to ~/.kiro/skills/ (global) or .kiro/skills/ (project-level). Each skill is a directory containing a SKILL.md file and optionally a references/ subdirectory with additional context the agent reads from the local filesystem when needed. Kiro discovers installed skills automatically and activates them on demand when a task matches.
Prerequisites: You need uv installed. An AWS account with credentials configured locally is required for API calls and script execution, but not for documentation search or skill discovery. See the user guide for detailed setup instructions.
fx
fx setup has the same two independent parts as Kiro: the AWS MCP Server (for runtime AWS API access and documentation search) and local skills (for task-specific agent guidance). Skills don't require the MCP server, and the MCP server doesn't serve locally-installed skills.
1. Add the AWS MCP Server from your terminal. fx stores MCP servers in ~/.fx/mcp.json, so the server is available in every project:
fx mcp add aws uvx mcp-proxy-for-aws-cli@latest https://aws-mcp.us-east-1.api.aws/mcp --metadata AWS_REGION=us-west-2
Or add it to ~/.fx/mcp.json by hand. fx uses the mcp key rather than mcpServers, type to select the transport, and a single command array holding the executable and its arguments:
{
"mcp": {
"aws": {
"type": "local",
"command": [
"uvx",
"mcp-proxy-for-aws-cli@latest",
"https://aws-mcp.us-east-1.api.aws/mcp",
"--metadata",
"AWS_REGION=us-west-2"
]
}
}
}
If a session is already open when you edit the file, run /mcp reload to pick up the change. The MCP server gives your agent access to AWS APIs, sandboxed script execution, and real-time documentation search.
2. Install skills from this repository:
npx skills add aws/agent-toolkit-for-aws/skills -a fx
This installs skill files to .fx/skills/ (project-level); add -g to install to ~/.fx/skills/ (global). fx also discovers skills already installed for other agents in .agents/skills/, .claude/skills/, and .codex/skills/, so a project set up for Codex or Claude Code needs no second copy. Each skill is a directory containing a SKILL.md file and optionally a references/ subdirectory with additional context the agent reads from the local filesystem when needed. fx reads skill metadata at startup and loads a skill's instructions when you or the agent invokes it; type $ in the composer or run /skills to browse what's installed.
Prerequisites: You need uv installed. An AWS account with credentials configured locally is required for API calls and script execution, but not for documentation search or skill discovery. See the user guide for detailed setup instructions.
Other agents
See the AWS MCP Server getting started guide for instructions on configuring the AWS MCP Server with your agent.
Then install skills from this repository:
npx skills add aws/agent-toolkit-for-aws/skills
Prerequisites: You need uv installed. An AWS account with credentials configured locally is required for API calls and script execution, but not for documentation search or skill discovery. See the user guide for detailed setup instructions.
What's included
Plugins
Plugins bundle the AWS MCP Server configuration and agent skills into a single install for your coding agent.
| Plugin | Description |
|---|---|
| aws-core | Core AWS skills and MCP Server configuration. Covers service selection, CDK/CloudFormation, serverless, containers, storage, observability, billing, SDK usage, and deployment. Start here. |
| aws-agents | Skills for building AI agents on AWS with Amazon Bedrock and AgentCore. |
| aws-data-analytics | Skills for data lake, analytics, and ETL workflows with S3 Tables, AWS Glue, and Athena. |
| aws-agents-for-devsecops | Investigate incidents, review code and execute UAT for release readiness, scan code for vulnerabilities, and run penetration tests with AWS DevOps Agent and AWS Security Agent. |
| aws-continuum | Discover, prioritize, validate, and remediate security risks across the software development lifecycle with AWS Continuum, right from your coding agent. See which risks are real and drive each one to a tested fix. |
| aws-startup-advisor | Puts an AWS Startups Solutions Architect in your coding assistant. Recommends the smallest architecture for your stage, credits, and team, then can scaffold it, migrate you from Azure, Google Cloud, or Heroku, move model calls to Amazon Bedrock, investigate production incidents, answer AWS Activate credits eligibility questions, and find startup offers you can use. |
Plugins are currently available for Claude Code, Codex, Cursor, and Devin. For other agents, configure the AWS MCP Server directly and install skills from this repository.
Skills
Agent skills are curated packages of instructions and reference materials that help agents complete specific AWS tasks. Skills are loaded on demand — agents discover and retrieve only what's relevant to the current task.
npx skills add aws/agent-toolkit-for-aws/skills
Browse the skills/ directory to see all available skills.
Rules files
Recommended project-level configuration files that tell agents how to use AWS most effectively — for example, by using the AWS MCP Server, discovering available skills, or searching documentation before acting.
See rules/ for details.
AWS MCP Server
The AWS MCP Server is a managed server that gives agents access to AWS through the Model Context Protocol. It provides:
- Full AWS API coverage — Interact with any of the 300+ AWS services through a single authenticated endpoint.
- Sandboxed script execution — Agents can run Python scripts in an isolated environment for complex multi-step operations.
- Real-time documentation access — Search and retrieve current AWS documentation, API references, and service capabilities without authentication.
- Enterprise controls — Amazon CloudWatch metrics, IAM context keys for agent-specific policies, and AWS CloudTrail audit logging.
For details on operation, available tools, authentication, and supported Regions, see the AWS MCP Server documentation.
Documentation
- User guide — Setup, configuration, and reference documentation.
- AWS MCP Server tools — Reference for all available MCP tools.
How the Agent Toolkit relates to the MCP servers, skills, and plugins in AWS Labs
In 2025, AWS began releasing MCP servers, skills, and plugins as part of AWS Labs. The Agent Toolkit for AWS is the successor to those tools. We recommend using the Agent Toolkit for AWS, because it offers key features including:
- IAM condition keys that distinguish between agent actions and human actions, so you can write policies that apply only to agents. For example, you can write policies that only allow read-only actions through the MCP server, even if the user’s underlying IAM role can take write actions).
- CloudWatch metrics and CloudTrail audit logging for every request, so you can monitor and audit coding agent activity.
- Agent skills that have undergone thorough end-to-end evaluations, so you can be confident that workflows will complete successfully.
AWS Labs MCP servers, skills, and plugins will continue to work and accept contributions, and over time the best of AWS Labs will be transitioned to the Agent Toolkit for AWS to ensure that customers can access the broadest array of tooling and guidance for their agents.
License
This project is licensed under the Apache-2.0 License. See LICENSE for details.