aws/agent-toolkit-for-aws

Official, AWS-supported MCP servers, skills, and plugins to help AI agents build on AWS

What it solves

It provides AI coding agents with the necessary tools, knowledge, and guardrails to build, deploy, and manage applications on AWS. It bridges the gap between general-purpose AI agents and the specific complexities of AWS services, ensuring agents can interact with the cloud environment safely and effectively.

How it works

The toolkit operates through three primary components:

  1. AWS MCP Server: A managed server using the Model Context Protocol (MCP) that gives agents access to 300+ AWS APIs, sandboxed Python script execution, and real-time documentation search.
  2. Agent Skills: Curated packages of instructions and reference materials that agents load on demand to complete specific AWS tasks.
  3. Plugins: Bundled packages (available for tools like Claude Code, Cursor, and Codex) that combine the MCP Server configuration and skills for easy installation.

Who it’s for

Developers using AI coding agents (such as Claude Code, Cursor, Codex, Kiro, or fx) who want to automate AWS infrastructure management, application deployment, and cloud architecture tasks.

Highlights

  • Broad Integration: Works with popular AI coding agents and the AWS CLI.
  • Comprehensive Coverage: Includes specialized plugins for core AWS services, AI agent building (Bedrock), data analytics, DevSecOps, and startup-specific architecture guidance.
  • Enterprise Guardrails: Features IAM condition keys to distinguish agent actions from human actions, along with CloudWatch metrics and CloudTrail audit logging.
  • Sandboxed Execution: Allows agents to run Python scripts in an isolated environment for complex operations.

Related

  • Project
    awslabs/agent-plugins

    A collection of plugins for AI coding agents that provide specialized skills, MCP servers, and workflows to help developers architect, deploy, and operate on AWS.

  • Project
    Community-Access/accessibility-agents

    A set of agent skills and tools that enforce WCAG 2.2 AA accessibility standards for AI coding agents across multiple platforms.

  • Project
    matlab/matlab-agentic-toolkit

    The MATLAB Agentic Toolkit is a MathWorks add‑on that connects LLM coding assistants (Claude Code, Copilot, Codex, Gemini, etc.) to a running MATLAB session via a local MCP server. It installs the server, loads curated “skills” covering core MATLAB and dozens of toolboxes, and exposes tools (run code, run tests, static analysis, list toolboxes) that the AI can call. This reduces hallucinations and token use, letting agents generate, test, and debug MATLAB code reliably.

  • Project
    cloudflare/skills

    A collection of agent skills and MCP servers that equip AI coding assistants with specialized knowledge of the Cloudflare Developer Platform for better guidance and deployment.

  • Project
    astronomer/agents

    AI agent tooling for data engineering that extends AI coding agents with specialized skills and an MCP server for managing Apache Airflow and querying data warehouses.