Frontend Design Without AI Slop

API contract review — branch feat/orders-v2 vs main

| # | Change | Endpoint | Severity | Rule |
|---|---|---|---|---|
| 1 | `customer_id` removed from response | GET /api/v1/orders/{id} | BREAKING | removed response field (R2) |
| 2 | `status` enum gains `refunded` | GET /api/v1/orders | ADDITIVE, risky | new enum value; strict clients may throw (A4) |
| 3 | `limit` max lowered 100 → 50 | GET /api/v1/orders | BREAKING | tightened validation (R5) |
| 4 | `X-Request-Id` header documented | all | COSMETIC | docs only |

Verdict: 2 breaking changes. Mobile 2.3.x reads `customer_id` on the order screen and will render an empty name.

Recommendation: keep `customer_id` as a deprecated alias for two releases, raise `limit` back to 100 with a server-side cap, and ship `refunded` behind the existing `status` field with a changelog note. No new API version needed if both are done.

Migration note for client teams: handle unknown `status` values as `other` before 2026-12-01.

Project: fresh `npm init -y` + `npm i -D vitest typescript` (Vitest 5.1.0, TypeScript strict). Every runner output below is pasted from the actual run.

Behaviour list (simplest first): 1. returns empty for blank input · 2. returns the address trimmed and lowercased when valid · 3. returns malformed when there is no @ · 4. returns malformed when there is more than one @ · 5. returns malformed when the domain has no dot · 6. returns too_long when the address is over 254 characters

Cycle 1: returns empty for blank input

```
[TDD] cycle 1/6  step RED  behaviour: returns empty for blank input
test:    src/validateEmail.test.ts > validateEmail > returns empty for blank input
ran:     npx vitest run src/validateEmail.test.ts -t "returns empty for blank input"
result:  FAIL  Error: Cannot find module './validateEmail' imported from <project>/src/validateEmail.test.ts
reason:  right (the unit under test does not exist yet: missing function, not a typo in the test)
changed: src/validateEmail.test.ts (+8, new file)
next:    minimum code in src/validateEmail.ts
```

```
[TDD] cycle 1/6  step GREEN  behaviour: returns empty for blank input
ran:     npx vitest run src/validateEmail.test.ts -t "returns empty for blank input" --reporter=verbose; npx vitest run src/validateEmail.test.ts
result:  PASS  Tests 1 passed (1)
changed: src/validateEmail.ts (+9, new file: result types + hard-coded `{ ok: false, error: 'empty' }`)
next:    refactor
```

Refactor: no refactor (hard-coded return is forced into the general version by cycle 2).

[Cycles 2–6 follow the same shape: each RED shows a real assertion mismatch such as `Tests 1 failed | 1 skipped (2)`, each GREEN the passing count, with refactors in cycles 5 and 6.]

Done

```
[TDD] DONE  cycles 6  tests added 6  suite PASS
ran:     npm test   (vitest run)   and   npx tsc -p .
result:  Test Files  1 passed (1) | Tests  6 passed (6); tsc exit 0
pre-existing failures: none (no tests existed)
found, not fixed: the 254 boundary itself (exactly 254 chars accepted) has no test; it passes today, so a test would be green on first run. 'a@.com' and 'a@b.' are accepted because the spec only asks for a dot in the domain.
```

No-fake-green check before every GREEN: no `.skip`/`.todo`, no assertion weakened, no test-only branch, nothing mocked.

> Assumption: signed-in users read their own orders and create their own orders as `pending`; every status change after that is made by your server with the `service_role` key, never by the browser.

Audit: `public.orders` (about 1.8M rows)

Rule files applied: `security-rls-basics`, `security-rls-performance`, `security-privileges`, `schema-foreign-key-indexes`, `query-missing-indexes`, `query-composite-indexes`, `schema-data-types`, `schema-constraints`, `lock-short-transactions`, `monitor-explain-analyze`.

Findings, by rule (first 6 of 12)

| # | Severity | Rule | Finding |
|---|---|---|---|
| 1 | CRITICAL | `security-rls-basics` | `orders_insert ... with check (true)` lets anyone who can insert write an order for any `user_id`. Combined with finding 2, an anonymous visitor holding only your public anon key can create orders on behalf of any user. |
| 2 | CRITICAL | `security-privileges` | `grant select, insert, update ... to anon`. Orders belong to signed-in users, so `anon` needs no privilege at all. None of the three policies has a `to` clause, so each applies to every role, including `anon`. |
| 3 | HIGH | `security-privileges` | `authenticated` can `update` every column of its own rows: the owner can still run `update orders set status = 'paid', total_cents = 1` from the browser. |
| 4 | HIGH | `security-rls-basics` | The insert path also trusts `status`: a client can insert a row that is already `status = 'paid'`. |
| 5 | HIGH | `security-rls-performance` | All policies call `auth.uid()` bare, so it is evaluated per row. Wrap it as `(select auth.uid())` so it is evaluated once per statement. |
| 6 | HIGH | `schema-foreign-key-indexes`, `query-missing-indexes` | No index on `user_id`. It is the foreign key, the RLS filter column and the column every "my orders" query filters on. Each RLS-filtered select over 1.8M rows is a sequential scan. |

Fixing migration (file 1 of 2, excerpt)

```sql
set local lock_timeout = '5s';

revoke all on public.orders from anon;
revoke update, delete, truncate, references, trigger on public.orders from authenticated;
grant select, insert on public.orders to authenticated;

drop policy if exists orders_select on public.orders;
drop policy if exists orders_insert on public.orders;
drop policy if exists orders_update on public.orders;

create policy orders_select on public.orders
  for select to authenticated
  using ((select auth.uid()) = user_id);

create policy orders_insert on public.orders
  for insert to authenticated
  with check ((select auth.uid()) = user_id and status = 'pending');
-- No update or delete policy: denied for every role except service_role.
```

File 2 holds only `create index concurrently if not exists orders_user_id_created_at_idx on public.orders (user_id, created_at desc);` because `concurrently` cannot run inside the transaction the Supabase CLI wraps around each migration.

[The full answer continues with findings 7–12, the lock plan for 1.8M rows, the status and total_cents check constraints added `not valid` then validated, an `explain analyze` to run before and after, and one RLS test per role.]

Pack: `references/pack-product-shots.md`.

Three options

1. PS-09 Strawberry Milkshake Product and Infographic. A ready-made premium cold-drink hero shot: seamless pastel backdrop, realistic condensation droplets and "premium café advertisement" styling, which is exactly the look a chilled cold-brew can needs.
Prompt by @Strength04_X (𝐌), source: https://x.com/Strength04_X/status/2090665058980462697, via awesome-nano-banana-pro-prompts, CC-BY-4.0.

2. PS-02 Luxury Chocolate Ad Still Life. The only prompt in the pack that is a full ad layout with on-image copy, but you would have to delete its placeholder copy and invented logos and rebuild most of the props.
Prompt by @AiwithLucas_ (Lucas AI), source: https://x.com/AiwithLucas_/status/2089231817916420518, via awesome-nano-banana-pro-prompts, CC-BY-4.0.

3. PS-01 Premium Beverage Advertising Photography. Written for a beverage-brand hero ad and works from your attached references, but the prompt is only one line long, so lighting and composition would come from what you attach.
Prompt by @exywuu (Exywuu), source: https://x.com/exywuu/status/2092222318084788564, via awesome-nano-banana-pro-prompts, CC-BY-4.0.

Pick: PS-09

What to attach. Image 1: a clear, front-on photo of your real Northbound can, logo facing the camera.

Adapted prompt (4:5 Instagram ad)

Image 1: product photo of the Northbound cold-brew coffee can. Use this exact can.

A vertical 4:5 Instagram feed ad. A high-quality professional product photograph of a premium cold-brew coffee can, the 330 ml matte navy aluminium can from Image 1 with its white logo, standing upright and centered against a soft pastel pink seamless studio background. Preserve the can's exact shape, proportions, matte navy finish, white logo and label from Image 1; do not redesign it, recolour it or add any new graphics to it. The can sits on the same seamless surface with a few clear ice cubes and a small scatter of dark roasted coffee beans at its base. Realistic condensation droplets cover the can with soft reflections from the studio lights. Elegant soft lighting creates subtle natural shadows beneath the can. The can fills about 55% of the frame height in the lower two thirds of the image, logo facing the camera.

Text: the headline "Cold. Bold. Northbound." across the top third of the image, centered, in a bold clean white sans-serif with generous letter spacing, large enough to read on a phone. Render this headline exactly once, spelled exactly as written. Apart from the can's own logo from Image 1, no other text, no prices, no taglines, no extra logos, no watermark.

Ultra-sharp focus, DSLR macro beverage photography, premium café advertisement style, hyper realistic, clean commercial composition, 8K.

Tool settings: aspect ratio 4:5; in AI Studio or the API also set the output resolution to 2K so the headline stays sharp.

Prompt by @Strength04_X (𝐌), source: https://x.com/Strength04_X/status/2090665058980462697, via awesome-nano-banana-pro-prompts, CC-BY-4.0. Adapted: subject swapped from a strawberry milkshake in a glass to the Northbound can, Image 1 reference and preserve-the-product line added, 4:5 aspect ratio and headline added, the original's second (infographic) paragraph removed.

[The full answer ends with a one-fix-per-turn table for headline contrast, product drift, garbled text and framing.]

Assumption: no n8n-MCP server is connected in this session, so I took the skill's fallback path: I wrote the workflow JSON directly from the bundled rules and checked it by review (a rule-check script plus a dry run of every expression). The real validator (`validate_workflow`) was not run, and nothing was deployed.

How the skill routed this

| Step | Sub-skill used | What it decided |
|---|---|---|
| Shape | `n8n-workflow-patterns` | Webhook processing pattern: Webhook, then Validate, then Action, then Respond |
| Node parameters | `n8n-node-configuration` | Slack post needs a channel and text. `responseMode: "responseNode"`. `options.responseCode` set on every Respond node. UUID node ids, no placeholder credentials |
| Expressions | `n8n-expression-syntax` | Webhook data lives under `.body`. Use `$('Webhook').item.json.body.*` once past the IF node. `=` prefix on every expression field |
| Error paths | `n8n-error-handling` | Slack gets `onError: "continueErrorOutput"` and its `main[1]` is wired to the 500 responder. `retryOnFail` is on. Bad input is caught upstream and gets a 400 |
| Validation | `n8n-validation-expert` | Runtime-profile checks done by review. Security warnings get fixed (header auth on the webhook) |

```
Webhook ──> Validate Input ──true──> Post to #signups ──success (main[0])──> Respond 200 OK
                        └──false──> Respond 400 Invalid Input
                                     Post to #signups ──error (main[1])────> Respond 500 Slack Failed
```

Validation result (excerpt)

```
PASS  Webhook responseMode is responseNode (Respond nodes control the reply)
PASS  Every branch (success and each error) ends at a Respond to Webhook node: no hanging branch
PASS  'Post to #signups': onError=continueErrorOutput AND main[1] wired
PASS  'Post to #signups': retryOnFail with maxTries <= 5 and waitBetweenTries <= 5000
PASS  Every {{ }} field starts with = (no 'Expression requires = prefix' error)
PASS  Webhook fields read from .body, never $json.name / $json.email / $json.plan
…
43 passed, 0 failed
```

Expression dry run against sample webhook items:

```
valid           IF -> true  (Slack)  Slack text: New signup: Ana Lima (ana@example.com) on pro
missing_plan    IF -> false (400)
bad_email       IF -> false (400)
no_body         IF -> false (400)
500 body:   {"error":"upstream_error","message":"The signup was received but could not be posted to Slack. Please retry later."}
```

Not verified without the server: the node typeVersions (Webhook 2.2, IF 2.3, Slack 2.3, Respond 1.5) and the Slack v2 parameter names; confirm by opening the nodes after import.

[The full answer also holds the six-node list, the importable workflow JSON and the expressions table with the reason behind each one.]

Template used: #2 YouTube thumbnail from `templates/use-cases.md`, with one extra line for the chart.

Defaults I chose, and why
- No numbers on the chart. The only figure in your request is 60%. If I asked for axis labels or dollar amounts, the model would make them up. The headline carries the number; the bars just show the drop.
- The "-60%" is spelled out. A minus sign next to a number is a common place for text errors, so the prompt names each character.
- `quality="high"`. The template defaults to `medium`; close-up faces and exact text both call for high.

Final prompt

A 16:9 YouTube thumbnail for a video titled "How I cut our AWS bill by 60%".

Scene: a dark, near-black navy studio backdrop, softly blurred, with a faint vignette and a subtle cool rim light. Simple and uncluttered, high contrast against the subject.

Subject: a photorealistic man in his early-to-mid 30s, fictional (not a real person), short dark hair, light stubble, plain charcoal crew-neck t-shirt. He fills the LEFT half of the frame, framed from mid-chest up and large in frame, head and shoulders slightly turned toward the right side. His expression is genuine surprise: eyebrows raised high, eyes wide, mouth open in an "O". One hand is raised near his cheek, palm open, fingers natural. He is looking toward the chart on the right, not at the camera. Sharp focus, bright key light on his face, real skin texture.

Chart: on the RIGHT half, a bold, simple bar chart of 5 glossy red (#E53935) bars that step down sharply from left to right, the tallest bar on the left and the shortest on the right. A thick red arrow runs diagonally downward along the tops of the bars, ending in an arrowhead at the bottom right. Flat, clean graphic style with a soft glow. No axis labels, no numbers, no gridlines and no legend on the chart.

Text (render exactly once, verbatim, no extra characters): "AWS BILL -60%". Spelled A-W-S, space, B-I-L-L, space, minus sign, 6, 0, percent sign. Heavy, condensed, all-caps sans-serif in bright yellow (#FFD400) with a thick black outline and a slight drop shadow. Placed on the RIGHT half, above the bar chart, the text block filling about one third of the frame height, left-aligned, never overlapping the man's face.

Style: bold, saturated, high-contrast YouTube thumbnail, readable at phone size. Yellow and red pop against the dark background.

Constraints: no other text anywhere, no AWS or Amazon logo, no other logos, no borders, no watermark, no emoji. Keep the bottom-right corner free of key detail (YouTube's timestamp sits there).

Settings: `model="gpt-image-2"`, `size="1536x864"`, `quality="high"`, `n=2`.

[The full answer also shows the filled template table, the Python call, and a one-change-per-turn fix table with a preserve list.]

2. Tool design

| Tool | Does | Inputs (constraints) | Returns | Read-only / destructive | Errors the model can fix |
|---|---|---|---|---|---|
| `find_customers` | Customers whose name or email contains the text, case-insensitive | `query` string 1–100 (trimmed), `limit` int 1–50 = 20, `cursor` optional | id, name, email; `next_cursor` | read-only, idempotent | no match: try a shorter part of the name or the email domain |
| `list_customer_orders` | One customer's orders, newest first | `customer_id` int ≥ 1, `status` enum optional, `limit`, `cursor` | customer + orders; `next_cursor` | read-only, idempotent | unknown customer: call `find_customers` |
| `get_order` | One order with the customer's name and email | `order_id` int ≥ 1 | order + customer (structured) | read-only, idempotent | not found: check the number or list the customer's orders |
| `refund_order` | Sets a paid order to refunded | `order_id` int ≥ 1, `confirm` int optional, must equal `order_id` | updated order, `previous_status: "paid"` | destructive, idempotent | no `confirm`: preview, ask the user; order pending or already refunded; not found |

Checklist notes: one rule for empty results (an `isError` result with a next step), applied to every tool. All SQL is parameterized. Read tools use a read-only SQLite handle; only `refund_order` opens a writable one, and its `UPDATE ... WHERE id = ? AND status = 'paid'` is a single statement. The guard on the destructive tool is server-side (the preview/confirm step), not just the `destructiveHint` annotation.

3. Scaffold

From `templates/typescript-server.md`. Installed: `@modelcontextprotocol/server` 2.4.0, `@modelcontextprotocol/client` 2.4.0, `zod` 4.6.7; Node 24.6.0.

```
build-an-mcp-server/
  data/seed.sql, data/shop.db     3 customers, 5 orders
  src/index.ts                    stdio entry point
  src/server.ts                   createServer(): the four tools
  src/shop.ts                     SQLite client module (the system being wrapped)
  test/smoke.ts                   spawns the server over stdio, calls every tool
```

5. Test (Inspector CLI, eval loop layer 2)

```
$ npx @modelcontextprotocol/inspector --cli npx tsx src/index.ts --method tools/call --tool-name get_order --tool-arg order_id=1001
{
  "content": [{ "type": "text", "text": "Order #1001 for Ada Lovelace <ada@example.com>: $49.99, paid, created 2026-10-02T09:15:00Z" }],
  "structuredContent": { "order": { "id": 1001, "customer_id": 1, "total_cents": 4999, "status": "paid", "created_at": "2026-10-02T09:15:00Z", "customer_name": "Ada Lovelace", "customer_email": "ada@example.com" } }
}
```

Built, tested, not tested

- Built: four tools over `data/shop.db`, server-side preview/confirm guard on `refund_order`, smoke test, configs for Claude Code, Codex and Cursor, README.
- Tested: starter smoke (pass), full smoke test 14 checks (pass), strict typecheck (pass), Inspector CLI call (pass), seed database unchanged after tests.
- Not tested: the Inspector web UI by hand and the model check (running the five real requests through a live Claude Code session and recording which tool it picks).

[The full answer also holds `src/shop.ts`, `src/server.ts`, `test/smoke.ts` and the three client config snippets.]

Phase 1: Root cause investigation
Full suite: FAIL test_expired_coupon_rejected_twice (tests/test_checkout.py:16) AssertionError: Coupon(code='OLD50', …expires_at=2020-01-01…) is not None.
In isolation the same test fails at a different line (17): AssertionError: 50.0 != 100.0. So the first lookup in a process is rejected and every later lookup is accepted; in the full suite the first lookup was done by test_expired_coupon_rejected. Cross-test pollution through module-level state.
Cache probe at the boundary: cache before [] → 1st None → cache after ['OLD50'] → 2nd Coupon(OLD50). Docstring says the cache was added in 3f2a1c9 "to cut DB reads": that is the regression point.
Trace (root-cause-tracing.md): apply_coupon → get_coupon → `if code in _CACHE: return _CACHE[code]`. get_coupon writes the coupon to _CACHE before the expiry check, and the cache-hit branch has no expiry check at all.

Phase 2: Pattern analysis
Working path: cache miss → _load → expiry check. Broken path: cache hit. Expiry depends on `now`, passed per call, so it cannot be a cached property; "don't cache expired coupons" alone would still serve a coupon cached while valid after it expires.

Phase 3: Hypothesis
"get_coupon returns cached coupons without re-checking expires_at against the caller's now; the cache was added around the lookup but the validity check stayed only on the miss path." Confirmed by the probe above.

Phase 4: Implementation
Failing test first: test_cached_coupon_rechecked_for_expiry (look up while valid, then after expiry) plus a setUp that clears the module cache. Before the fix: 5 tests, 2 failures, both for the expected reason.
Fix at the source: the cache stores the raw record; the expiry check runs on every lookup, cached or not.

    c = _CACHE.get(code)
    if c is None:
        c = _load(code)
        if c is None:
            return None
        _CACHE[code] = c
    if c.expires_at < now:
        return None
    return c

Verification: python3 -m unittest -v → Ran 5 tests … OK. The originally failing test in isolation, 3 runs: OK, OK, OK. Fix attempts: 1.