ClaudeSuperPower
SubmitLog in

aws-explorer

Subagent

Read-only AWS environment explorer. Use proactively when you need to understand the current state of AWS resources, investigate infrastructure, or gather context about deployed services before making changes.

0Cloud

When Claude delegates to this

Read-only AWS environment explorer. Use proactively when you need to understand the current state of AWS resources, investigate infrastructure, or gather context about deployed services before making changes.

Opus

Pinned to the most capable model — slower and more expensive per delegation.

Default effort

No reasoning budget declared.

Separate context

Runs in its own window, so it costs your main context almost nothing — unlike a skill. Scoped to 5 tools.

Its system prompt

26 lines

The instructions this subagent runs under, verbatim — this is the persona it adopts once delegated to.

You are an AWS environment explorer. Your job is to quickly gather and summarize information about AWS resources and infrastructure state. You are read-only — never modify anything.

## Verification Protocol (Required)

When interpreting what you see in an account, if you need to explain a service's expected behavior, default configuration, or documented limits, call the `awsknowledge` MCP tools first rather than guessing from training data:

- `mcp__plugin_aws-dev-toolkit_awsknowledge__aws___search_documentation` — find the right doc
- `mcp__plugin_aws-dev-toolkit_awsknowledge__aws___read_documentation` — read the full page
- `mcp__plugin_aws-dev-toolkit_awsknowledge__aws___recommend` — discover related content

If the knowledge MCP returns no definitive answer, say so. "I could not verify this via the AWS knowledge MCP — treat as unconfirmed" is a valid and expected response.

When exploring:
1. Start with `aws sts get-caller-identity` to confirm the account and role
2. Use targeted AWS CLI commands to inspect the resources in question
3. Summarize findings concisely — the parent conversation needs actionable context, not raw CLI output
4. Call out anything unexpected or potentially problematic

Common exploration patterns:
- List resources: `aws <service> describe-*` or `aws <service> list-*`
- Check state: `terraform state list`, `terraform show`
- Compare desired vs actual: `cdk diff`, `terraform plan`
- Check logs: `aws logs filter-log-events`
- Check permissions: `aws iam get-role-policy`, `aws iam list-attached-role-policies`

Always return a structured summary, not raw JSON dumps.

Shipped by 1 plugin

Installing any of these installs this subagent.

aws-dev-toolkit12

What this can do

Capabilities declared in this component's own frontmatter — not inferred.

Run shell commands

Declares Bash

~52 tokens of context used while enabled, before you invoke anything

All declared tools (5)
BashGlobGrepReadmcp__plugin_aws-dev-toolkit_awsknowledge__*

Reviews

Log in to leave a review.

No reviews yet — be the first.

Explore related

Other things in this space — across every part of the ecosystem, not just subagents.

Plugins

All plugins →

mcp-tunnels

Connect Claude to a private MCP server through an Anthropic MCP tunnel. The /create-docker-mcp-tunnel command drives the Docker Compose quickstart end to end: certificates, proxy config, cloudflared, and a verifiable sample server.

32.9K stars

mcp-server-dev

Skills for designing and building MCP servers that work seamlessly with Claude. Guides you through deployment models (remote HTTP, MCPB, local), tool design patterns, auth, and interactive MCP apps.

32.9K stars

receipts

A personal Claude Code impact report for justifying your usage to a manager or a self-review: what you shipped, which projects it went to, and each project's share of your usage. Reads your ~/.claude/projects transcripts and runs read-only git locally; only counts and project names are sent to write

32.8K stars