Borrowing it
Nothing to install: this file belongs to cardstack/boxel. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/cardstack/boxel/main/.claude/skills/aws-access/SKILL.mdgit clone --depth 1 https://github.com/cardstack/boxelWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cardstack/boxel/aws-access)<a href="https://agentmods.dev/skills/cardstack/boxel/aws-access"><img src="https://agentmods.dev/badge/skills/cardstack/boxel/aws-access/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cardstack/boxel/aws-access"><img src="https://agentmods.dev/badge/skills/cardstack/boxel/aws-access.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.14383 |
| Opus 5.5 | $0.00000 | $0.05753 |
| Sonnet 5.5 | $0.00000 | $0.02877 |
| Haiku 4.5 | $0.00000 | $0.01438 |
Grade D, and why
aws-access scanned grade D with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- `session-manager-plugin` installed, **above 1.2.497.0** — required for the SSM port-forward tunnel to RDS. Install instructions: <https://docs.aws.amazon.com/systems-manager/latest/userguide/session-manager-working-wit Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
The flow below solves both: a wrapper script that mints a scoped STS session and writes its credentials to a _named profile_ in `~/.aws/credentials` so Claude can read them from any shell, plus the SSM port-forward patte Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Do not use `localhost` (or `127.0.0.1`) as the `host` parameter. The ECS-managed SSM agent refuses to forward to loopback addresses for `AWS-StartPortForwardingSessionToRemoteHost`: the local port still binds and the ses How it starts
The opening of the file, as written. The whole thing — 723 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AWS access (staging + prod) and RDS querying
This skill exists because:
- The team's existing
staging/prodshell aliases export AWS credentials into the user's interactive shell, but Claude's Bash tool spawns its own shell that does not inherit them. - The boxel staging/prod RDS instances are not publicly reachable — they live inside the AWS VPC.
The flow below solves both: a wrapper script that mints a scoped STS session and writes its credentials to a named profile in ~/.aws/credentials so Claude can read them from any shell, plus the SSM port-forward pattern for talking to the in-VPC database.
The script and mise task are in this repo:
scripts/claude-aws.shmise-tasks/claude-aws
Profile convention is fixed (don't change without a heads-up):
claude-staging— staging temp session (holdsboxel-claude-readonlyrole creds)claude-prod— prod temp session (holdsboxel-claude-readonlyrole creds)
Identity model (CS-10962)
The credentials in [claude-staging] / [claude-prod] are not the user's IAM identity. The script does:
- Reads the user's long-lived IAM access keys (held in their source profile, e.g.
cardstack). - Calls
sts:AssumeRole arn:aws:iam::<account>:role/boxel-claude-readonlydirectly, passing the user's MFA token via--serial-numberand--token-code. The role's trust policy requiresaws:MultiFactorAuthPresent: true, which is satisfied by these flags. This is the MFA gate. - Writes the role's credentials to
[claude-<env>]. Session length is the role'smax_session_duration(12h).
Net effect: every aws --profile claude-<env> ... call Claude makes runs as boxel-claude-readonly, with exactly that role's policy. The user's IAM group memberships only matter to the extent that they grant sts:AssumeRole on the role; once the role is assumed, the user's groups are no longer in the picture. The role is the same name (boxel-claude-readonly) in both staging and prod, provisioned by the infra side of CS-10962.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 723 lines · 0 tokens per session scan D 993e2e0e4414
aws-access is a skill published in the GitHub repository cardstack/boxel (27 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 14,383 tokens. A static security scan graded it D with 3 findings (asks for root, reaches for credential files, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-21.
Other skills, from other repositories
migrate-to-shoehorn
Migrate test files from as type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace as in tests, or needs partial test data.
typescript-test-writing
Use this skill when writing or modifying tests in the llxprt-code repository. Covers mandatory TDD, behavioral testing, bun:test conventions and file naming, mock hygiene (no mock theater), and what never to test. Distilled from dev-docs/RULES.md, which remains the source of truth.
typescript
TypeScript strict mode with eslint and jest.
nestjs
Use when building or structuring a NestJS backend — feature modules, providers and DI wiring, provider scopes and request-lifecycle order, where to bind guards/pipes/interceptors/filters, and testing with Test.createTestingModule. NOT a bare Express/Fastify service with no DI (that is nodejs), NOT framework-agnostic…
jest-unit
Unit testing skill using Jest for TypeScript and JavaScript, covering mocking, spies, snapshots, coverage, async testing, and custom matchers.
Jasmine Testing
BDD-style JavaScript testing with Jasmine covering spies, async patterns, custom matchers, clock manipulation, and comprehensive test organization for frontend and Node.js applications.