aws-auth

An agent skill by aws, from aws/agent-toolkit-for-aws. Tags: access-control, api, mobile, security, web.

What it does

Adds user authentication to web and mobile apps with Amazon Cognito (user pools and identity pools) and the AWS Amplify client auth libraries. Covers sign-up/sign-in flows and the login page (Cognito-hosted UI / managed login), MFA, password policies, OAuth 2.0 / OIDC flows (auth-code + PKCE, client credentials), social/SAML federation, tokens (ID/access/refresh, rotation, revocation, storage), Cognito Lambda triggers, identity pools (temp AWS creds), and gating API Gateway (or ALB) routes to signed-in users via Cognito/JWT authorizers. Applies when adding a login or sign-up page, configuring a user pool or app client, choosing user pool vs identity pool, wiring social/SAML, refreshing tokens, requiring sign-in on an API Gateway or ALB, or debugging redirect_uri/token/MFA/CORS/federation errors. Does NOT cover Amplify Gen2 backend definitions (defineAuth, npx ampx → aws-amplify), IAM/STS/Identity Center (→ aws-iam), or API Gateway/Lambda resource config beyond the authorizer (→ aws-serverless).

Install

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

npx skills add aws/agent-toolkit-for-aws --skill aws-auth

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-core/skills/aws-auth ~/.claude/skills/aws-auth

Safety box score

Grade C, moderate risk: 39 of 100 risk points, rated 2026-09-21. 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 configures Cognito user pools, app clients, identity pools, hosted-login domains, API Gateway authorizers, and related Lambda permissions. The main risks are broad AWS-side auth changes, full-replace update calls that can silently wipe settings, and handling federation secrets or account-confirmation shortcuts incorrectly.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	2 (moderate)	It drives shell-style AWS CLI and curl workflows such as `update-user-pool-client`, `set-identity-pool-roles`, `create-authorizer`, and `/oauth2/token`, so an agent can execute real commands against remote services.
Network access	2 (moderate)	It sends auth material and configuration to AWS endpoints (`/oauth2/token`, `/oauth2/revoke`, Cognito, API Gateway, Secrets Manager), so data regularly leaves the machine over the network.
Filesystem reach	1 (low)	The only local file use is limited and scoped, e.g. `mktemp`, `client.json`, and `client-updated.json`; it does not instruct broad reads or writes outside the working area.
Secrets & credentials	2 (moderate)	It handles client secrets, refresh tokens, session cookies, and secret-manager lookups (for example `client_secret` in `create-identity-provider`), so a mistake could expose credentials.
Privilege & persistence	0 (none)	There are no sudo/install-global/service/shell-profile operations; the changes are to AWS resources, not local machine privilege settings.
Instruction hijack surface	0 (none)	No hidden prompt-injection or instruction-override content is present; warnings like "Do NOT use" and "trust the docs over memory" are ordinary operational guidance.
Irreversible actions	3 (high)	It can create and modify live auth infrastructure (`create-user-pool-domain`, `update-user-pool-client`, `set-identity-pool-roles`, `create-authorizer`, `lambda add-permission`), which can alter sign-in and access paths across an AWS account without an explicit human gate.
Transparency	0 (none)	All shipped files are readable markdown with plain command snippets; there are no binaries, obfuscated blobs, or unpinned installers.

Before installing

Run it only with an explicit AWS profile/account you are willing to modify; these commands can change live Cognito/API Gateway/Lambda auth settings.
Read the full-replace warnings in `SKILL.md`, `references/managed-login-oauth.md`, `references/identity-pools.md`, and `references/threat-protection.md` before any `update-*` or `set-*` call.
If you use social or SAML federation, confirm the IdP secret will stay in Secrets Manager and never appear in shell history or logs.
Decide up front whether you are okay with any account-confirmation shortcuts (`admin-confirm-sign-up`) or whether you want only user-verified sign-ups.

Rated 2026-09-21 by an automated review of 10 files from the skill's repository.

Source

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

aws-secrets-manager — Secret safety for AWS Secrets Manager, secret management, credentials, API keys, tokens, and passwords.
csharp-developer — Use when building C# applications with .NET 8+, ASP.NET Core APIs, or Blazor web apps. Builds REST APIs using minimal or controller-based routing.
django-storages-s3 — Use when configuring Django to store static and media files on AWS S3 with django-storages.
dotnet-core-expert — Use when building .NET 8 applications with minimal APIs, clean architecture, or cloud-native microservices. Invoke for Entity Framework Core.
fullstack-guardian — Builds security-focused full-stack web applications by implementing integrated frontend and backend components with layered security at every level.
java-architect — Use when building, configuring, or debugging enterprise Java applications with Spring Boot 3.x, microservices.