Two systems — coupons and workflows — chained into one campaign.
I want a welcome programme for new customers. Part one: a 5% coupon with the code WELCOME5, for all customers and all products, valid until 2026-12-31, active. Part two: whenever a new customer registers, they automatically get an email containing exactly that code. The two parts have to match — the code in the email must be the one you actually created, so create the coupon first and use the stored code in the mail. Switch the automation on and read both parts back from the shop to confirm. Show me the email text you wrote.
Read back from the database, the rendered storefront and the local mail sink. No model involved.
| Result | Check | Weight |
|---|---|---|
| PASS | Coupon WELCOME5 exists in tkupon kKupon=2150 name="Willkommensgutschein" |
3 |
| PASS | WELCOME5 discounts 5% (cWertTyp=prozent) fWert=5 cWertTyp=prozent |
3 |
| PASS | WELCOME5 is active and valid through 2026-12-31 cAktiv=Y dGueltigBis=2026-12-31 23:59:00 |
2 |
| PASS | A workflow was created on a trigger matching customer/kunde/regist created: #3956 newCustomer → sendMail |
3 |
| PASS | That workflow is active isActive=1 |
2 |
| PASS | Its action is an e-mail action action=sendMail |
2 |
| PASS | The workflow email body contains the WELCOME5 code code embedded in workflow arguments |
3 |
Scored by opus, which saw only the evidence bundle.
| Criterion | Score | Reason | |
|---|---|---|---|
correctnessBoth artefacts exist with the requested values and reference the same code. |
9/10 | state_diff.json shows tkupon 2150 WELCOME5 5% prozent cAktiv=Y bis 2026-12-31 23:59 with no customer/group/article restriction, and workflow 3956 newCustomer→sendMail isActive=1 whose body carries the same literal WELCOME5 (calls.jsonl:16-17). | |
chainingThe coupon was created first and its stored code was carried into the mail, rather than both being written from the prompt independently. |
9/10 | Order on the wire is duplicate-check SQL (call 7) → dry-run → real create (call 10, kKupon 2150) → full read-back of tkupon.cCode (call 11) → only then create_workflow with that code (calls 15-16), so the mail code came from the stored row, not from the prompt alone. | |
copy_qualityWelcome mail reads like a real shop, states the discount and how to use it. |
7/10 | Mail names the 5% discount, the code, where to enter it ("an der Kasse") and the expiry, with {$customer->cVorname} personalisation and valid HTML, but signs off with a bare "Ihr Team" instead of the available {$global->shopName} and mentions no minimum-order or usage terms. | |
communicationReports both parts and how they are linked. |
8/10 | ANSWER.md gives a coupon table, the workflow id/trigger/action, quotes the raw DB read-back (kKupon 2150, cAktiv=Y, fWert=5, dGueltigBis) and the full mail text, and states explicitly that the mail carries the code from part one — though it asserts the link rather than showing the read-back that established it. |
It verified the code was free, dry-ran both writes, then read the stored coupon row back from tkupon before composing the mail — a genuine create-then-read-then-reuse chain rather than two independent writes from the prompt.
The email copy is competent but generic — no shop name (the newCustomer trigger exposes {$global->shopName}), no terms beyond the expiry date — and the answer asserts the code match rather than showing the read-back that proves it.
| Tool | Calls | Errors | Avg |
|---|---|---|---|
run_sql_read | 4 | 1 | 306 ms |
invoke_admin_action | 3 | 0 | 416 ms |
list_admin_actions | 2 | 0 | 425 ms |
describe_admin_action | 2 | 0 | 382 ms |
describe_workflow_capabilities | 2 | 0 | 295 ms |
create_workflow | 2 | 0 | 382 ms |
get_guide | 1 | 0 | 334 ms |
list_workflows | 1 | 0 | 286 ms |
get_workflow | 1 | 0 | 415 ms |
# Willkommensprogramm – eingerichtet
## 1. Der Gutschein
| Feld | Wert |
|---|---|
| Code | **WELCOME5** |
| Rabatt | 5 % |
| Gültig für | alle Kunden, alle Kundengruppen, alle Produkte |
| Gültig bis | 31.12.2026, 23:59 Uhr |
| Status | aktiv |
Bestätigt direkt aus der Datenbank (kKupon 2150): `cCode=WELCOME5, cAktiv=Y, fWert=5, dGueltigBis=2026-12-31 23:59:00`.
## 2. Die Automatisierung
Neuer Workflow **"Willkommensprogramm: Neukunden erhalten den 5%-Gutschein WELCOME5 per E-Mail"** (ID 3956), aktiv.
- Auslöser: Neue Kundenregistrierung
- Aktion: E-Mail senden, an die E-Mail-Adresse und den Namen des neuen Kunden
- Enthält exakt den oben angelegten Code WELCOME5
Bestätigt: Workflow ist aktiv und feuert (`fires: true`).
## Die E-Mail an neue Kunden
**Betreff:** Willkommen bei uns – 5% Rabatt auf Ihre erste Bestellung
> Hallo {Vorname},
>
> schön, dass Sie sich bei uns registriert haben!
>
> Als Willkommensgeschenk erhalten Sie **5% Rabatt** auf Ihre nächste Bestellung. Nutzen Sie einfach folgenden Gutscheincode an der Kasse:
>
> **WELCOME5**
>
> Der Gutschein ist gültig bis 31.12.2026 und gilt für alle Produkte.
>
> Wir wünschen Ihnen viel Freude beim Einkaufen!
>
> Ihr Team
Jede neue Registrierung löst diese Mail automatisch aus, mit dem echten Gutscheincode aus Teil 1.
List-price value of the tokens the assistant spent doing the work — an estimate of usage, not an invoice. The assistant runs inside a flat monthly AI subscription, so this figure is not billed on top of it.
Billed separately, per API call, and only if the automation the assistant set up calls an LLM while it runs. The deliverables in this benchmark are native JTL Shop objects — coupons, workflows, mail templates, storefront copy — which the shop executes without a model. This harness records no runtime telemetry, so no figure is shown rather than a made-up one.
Run 20260810-205005_f-welcome-program_sonnet · shop reset to fixture before the run · restore with jtl restore 20260810-205005_f-welcome-program_sonnet
© 2026 the author · scores are generated from recorded runs, not written by hand.