agents-get-started

An agent skill by aws, from aws/agent-toolkit-for-aws. Tags: architecture, cli, developer-tools, local-workflow, setup.

What it does

Use when a developer wants to create a new agent project or get started with AgentCore. Handles framework selection, project scaffolding, first deploy, and first invocation. Triggers on: "build an agent", "create an agent", "get started", "new project", "agentcore create", "which framework", "Strands vs LangGraph", "hello world agent", "first agent", "create MCP server", "host MCP server", "agentcore dev", "dev server", "what port", "local development". Not for adding capabilities to existing projects — use agents-build or agents-connect. Strands vs LangGraph in a migration context routes to agents-build, not here. Connecting to an existing MCP server routes to agents-connect, not here.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add aws/agent-toolkit-for-aws --skill agents-get-started

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/aws/agent-toolkit-for-aws
cp -r agent-toolkit-for-aws/plugins/aws-agents/skills/agents-get-started ~/.claude/skills/agents-get-started

Safety box score

Grade D, elevated risk: 49 of 100 risk points, rated 2026-10-19. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This skill scaffolds a new AgentCore project and walks the user through local dev, first deploy, and first invocation on AWS. The main risks are that it runs shell commands, may require a global CLI install, and can create live cloud resources and touch local env files.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	2 (moderate)	It explicitly uses Bash to run commands like `agentcore --version`, `agentcore create`, `agentcore dev`, and `agentcore deploy`, and falls back to `npm install -g @aws/agentcore` if the CLI is missing.
Network access	2 (moderate)	The skill sends the user toward AWS-hosted deployment (`agentcore deploy`) and remote model providers; the reference also shows local HTTP invocation with `curl -X POST http://localhost:8080/invocations`.
Filesystem reach	1 (low)	It scaffolds files inside the project tree (`agentcore/`, `app/<AgentName>/`) and writes `agentcore/.env.local`, but it does not direct edits or deletes outside the workspace.
Secrets & credentials	2 (moderate)	It explicitly uses `agentcore/.env.local` for local environment variables and says deploy writes resource IDs there; the reference also names provider keys like `ANTHROPIC_API_KEY`, `OPENAI_API_KEY`, and `GEMINI_API_KEY`.
Privilege & persistence	2 (moderate)	If the CLI is missing, it tells the user to run a global install: `npm install -g @aws/agentcore`, which affects the whole user environment.
Instruction hijack surface	0 (none)	No hidden instructions or prompt-injection behavior is present; instead, it tells the agent to ask before creating or deploying and not to proceed automatically.
Irreversible actions	3 (high)	Its core purpose is to create and first-deploy a new agent, including `agentcore deploy` and `agentcore invoke`, which can create live AWS resources and externally reachable endpoints.
Transparency	0 (none)	Only readable markdown files are shipped here; there are no binaries, minified blobs, or unpinned remote installers in the skill itself.

Before installing

Read the `agentcore deploy` and `agentcore/.env.local` sections first; confirm you’re okay with live AWS resources and local env files being created.
If you may use Anthropic/OpenAI/Gemini, decide how those API keys will be supplied before running the skill.
Try `agentcore create --dry-run` in a scratch account or sandbox first.
Be comfortable with a global install if the CLI is missing: `npm install -g @aws/agentcore`.

Rated 2026-10-19 by an automated review of 2 files from the skill's repository.

Source

Repository: aws/agent-toolkit-for-aws (all skills from this repository)
Path: plugins/aws-agents/skills/agents-get-started/SKILL.md
Branch: main
Collection: aws-agents
Updated: 2026-09-19

context7-cli — Use the ctx7 CLI to fetch library documentation, manage AI coding skills, and configure Context7 MCP.
dxf — Generate, regenerate, and validate 2D DXF drawings from Python build123d sources.
filesystem-context — This skill should be used when agent work needs file-backed context: durable scratchpads, tool-output offloading, just-in-time discovery.
setup-gbrain — Set up gbrain for this coding agent: install the CLI, initialize a local PGLite or Supabase brain, register MCP, capture per-remote trust policy. (gstack)
statusline-generator — Installs, configures, customizes, or troubleshoots the Claude Code statusline (cwd, model, token counts).
aws-blocks — Guides building full-stack applications with AWS Blocks — an Infrastructure-from-Code framework.