Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Validates AWS readiness for website deployment. Checks CLI tools, credentials, SES, Route 53, and ACM. Produces a report with pass/fail and action items.
.claude/skills/grcengclub-website-preflight/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 104% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 69% | 0% |
| case-13 | ✗→✓ | ▲ Improved | 5% | 0% |
You are running the /grc-portfolio:preflight skill. Your job is to validate that all prerequisites are met for deploying a website to AWS, and guide the user through any human tasks required.
Find site-config.json:
$ARGUMENTS for a project directory pathRead it to understand what features are enabled and what AWS configuration is needed.
Run each check and record the result as PASS or FAIL:
bash aws --version If missing: suggest running $TOOLKIT_DIR/scripts/bootstrap.sh
bash node --version
bash npm --version
bash aws sts get-caller-identity --profile <aws.profile> Record the account ID and ARN for the report.
Verify aws.region in config is us-east-1 (required for CloudFront + ACM).
Run $TOOLKIT_DIR/scripts/validate-stack.sh against the appropriate template:
features.customDomain: validate $TOOLKIT_DIR/cloudformation/website-infrastructure.yaml$TOOLKIT_DIR/cloudformation/website-infrastructure-no-domain.yamlfeatures.contactForm)bash aws ses get-identity-verification-attributes --identities <client.email> --profile <aws.profile> --region us-east-1 Check that the status is "Success".
features.customDomain)bash aws route53 list-hosted-zones-by-name --dns-name <aws.domain> --profile <aws.profile> If found, save the hosted zone ID to aws.hostedZoneId in config.
features.customDomain)bash aws acm list-certificates --region us-east-1 --profile <aws.profile> --query "CertificateSummaryList[?DomainName=='<aws.domain>']" If found and status is "ISSUED", save the ARN to aws.certArn in config.
bash gh --version
Output a formatted report with three sections:
List all checks that passed with a checkmark.
Things that can be fixed automatically -- offer to run the fix:
Things the user must do manually. For each, provide:
Common human tasks:
brew install gh && gh auth loginIf there are human action items, tell the user to complete them and then say "ready" or "done" to re-run the checks.
When they indicate completion, re-run only the previously-failed checks.
Once all checks pass:
aws.hostedZoneId if discoveredaws.certArn if discoveredstatus.preflightComplete = truesite-config.jsonTell the user:
/grc-portfolio:build next (if not done) or /grc-portfolio:infra to deploy infrastructure$TOOLKIT_DIR = read from site-config.json toolkitDir field$ARGUMENTS = arguments passed after /grc-portfolio:preflight (expected: project directory path)| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 4,699 | 11,374 | +142% | 1 | 1 | 0% | 443 | 1,423 | +221% | 0 | 0 | — |
case-02 | fail→fail | 5,972 | 5,322 | -11% | 1 | 1 | 0% | 303 | 1,516 | +400% | 0 | 0 | — |
case-03 | fail→fail | 12,548 | 5,300 | -58% | 1 | 1 | 0% | 2,343 | 1,444 | -38% | 0 | 0 | — |
case-04 | fail→fail | 4,619 | 3,822 | -17% | 1 | 1 | 0% | 798 | 1,739 | +118% | 0 | 0 | — |
case-05 | fail→fail | 5,863 | 8,537 | +46% | 1 | 1 | 0% | 964 | 1,913 | +98% | 0 | 0 | — |
case-06 | fail→fail | 11,056 | 27,669 | +150% | 1 | 1 | 0% | 2,041 | 1,547 | -24% | 0 | 0 | — |
case-07 | fail→pass | 17,722 | 5,250 | -70% | 1 | 1 | 0% | 967 | 1,972 | +104% | 0 | 0 | — |
case-08 | fail→pass | 10,056 | 1,728 | -83% | 1 | 1 | 0% | 1,660 | 1,407 | -15% | 0 | 0 | — |
case-09 | fail→pass | 11,143 | 14,566 | +31% | 1 | 1 | 0% | 1,734 | 1,485 | -14% | 0 | 0 | — |
case-10 | pass→pass | 15,963 | 5,137 | -68% | 1 | 1 | 0% | 2,697 | 2,048 | -24% | 0 | 0 | — |
case-11 | pass→pass | 9,355 | 2,058 | -78% | 1 | 1 | 0% | 1,419 | 1,442 | +2% | 0 | 0 | — |
case-12 | fail→pass | 6,507 | 2,857 | -56% | 1 | 1 | 0% | 995 | 1,679 | +69% | 0 | 0 | — |
case-13 | fail→pass | 11,423 | 3,665 | -68% | 1 | 1 | 0% | 1,785 | 1,867 | +5% | 0 | 0 | — |
case-14 | fail→pass | 12,860 | 5,243 | -59% | 1 | 1 | 0% | 2,085 | 2,098 | +1% | 0 | 0 | — |
case-15 | fail→pass | 10,547 | 5,427 | -49% | 1 | 1 | 0% | 1,705 | 2,153 | +26% | 0 | 0 | — |
case-16 | fail→pass | 9,230 | 3,908 | -58% | 1 | 1 | 0% | 1,356 | 1,914 | +41% | 0 | 0 | — |
case-17 | fail→pass | 7,283 | 3,316 | -54% | 1 | 1 | 0% | 1,188 | 1,710 | +44% | 0 | 0 | — |
case-18 | fail→pass | 10,388 | 3,356 | -68% | 1 | 1 | 0% | 1,653 | 1,701 | +3% | 0 | 0 | — |
case-19 | pass→pass | 5,602 | 1,764 | -69% | 1 | 1 | 0% | 799 | 1,396 | +75% | 0 | 0 | — |
case-20 | pass→pass | 8,252 | 1,599 | -81% | 1 | 1 | 0% | 1,288 | 1,416 | +10% | 0 | 0 | — |
case-21 | fail→pass | 14,008 | 1,486 | -89% | 1 | 1 | 0% | 2,093 | 1,345 | -36% | 0 | 0 | — |
case-22 | fail→pass | 7,036 | 7,834 | +11% | 1 | 1 | 0% | 1,051 | 2,084 | +98% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 18 counted toward the lift figure. The other 4 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +55 percentage points is the difference between those two pass rates over the 18 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.