@cloudgrid-io/mcp 0.12.0 → 0.14.0

package/src/corpus/templates/restaurant-reservations/AGENTS.md

@@ -0,0 +1,258 @@
+# AGENTS.md — restaurant-reservations (BLUEPRINT)
+
+This is a **structure guide**, not runnable code. It tells an agent how to build a
+restaurant website with online reservations correctly **on CloudGrid**: a public
+Next.js site (menu + booking form) backed by grid-shared MongoDB, with the
+CloudGrid-native wiring for the DB, secrets, optional payments/auth, and a
+scheduled reminder job. Read it end to end, then build the app under
+`services/web/` following the same proven shape as the `app-with-data` template.
+
+Do NOT ship this as-is — there is no app code here. Adapt the tree below.
+
+---
+
+## 1. File tree
+
+App code MUST live under `services/<service-name>/`. The service is named `web`,
+so the CLI looks for `services/web/`. `path: /` in `cloudgrid.yaml` is the URL
+mount, NOT the filesystem path — files at the repo root fail with
+`Error: Service directory not found: …/services/web`.
+
+```
+cloudgrid.yaml                                   # name + services.web (nextjs) + needs:{database:true} + vault:
+AGENTS.md                                        # this guide
+services/web/package.json                        # next, react, react-dom, mongodb (+ stripe if deposits)
+services/web/lib/db.js                           # lazy Mongo client from DATABASE_MONGODB_URL (legacy MONGODB_URL fallback)
+services/web/lib/mail.js                         # SendGrid client from process.env.SENDGRID_API_KEY (vault-injected)
+services/web/app/layout.js                       # root layout + site chrome / inline CSS
+services/web/app/page.js                         # home: hero + menu (server component, reads menu collection)
+services/web/app/reserve/page.js                 # reservation page (renders the booking form)
+services/web/app/reserve/reservation-form.js     # "use client" form: POST a reservation
+services/web/app/api/menu/route.js               # GET menu items (public)
+services/web/app/api/reservations/route.js       # POST create reservation, GET list (staff)
+services/web/app/api/reservations/[id]/route.js  # PATCH status (confirm/seat/cancel), DELETE
+# ─ optional: card deposit to hold a table (Stripe) ─
+services/web/app/api/checkout/route.js           # POST → create Stripe Checkout Session for a deposit
+services/web/app/api/webhooks/stripe/route.js    # POST ← Stripe webhook: mark reservation deposit paid
+# ─ optional: staff auth (Clerk/Auth0) ─
+services/web/middleware.js                        # protect /staff/* routes with the provider SDK
+services/web/app/staff/page.js                    # staff dashboard: today's reservations
+# ─ scheduled reminders (SEE §7 — cron deploy PENDING #1543) ─
+services/reminder/package.json                    # mongodb + sendgrid; a one-shot job entry
+services/reminder/index.js                        # find tomorrow's reservations → send reminder emails
+```
+
+Keep the public reservation flow **unauthenticated** — a diner should book
+without an account. Auth is only for the staff dashboard.
+
+---
+
+## 2. MongoDB collections + fields
+
+The grid provisions one shared Mongo database (see §3). Use these collections:
+
+### `menu`
+Menu content shown on the home page. Seed it once (script or an admin route).
+| field | type | notes |
+|-------|------|-------|
+| `_id` | ObjectId | |
+| `section` | string | e.g. "Starters", "Mains", "Desserts" |
+| `name` | string | dish name |
+| `description` | string | short copy |
+| `price` | number | in cents, or a display string — pick one and be consistent |
+| `order` | number | sort order within a section |
+| `available` | boolean | hide sold-out items |
+
+### `reservations`
+The booking records — the persistent core of this app.
+| field | type | notes |
+|-------|------|-------|
+| `_id` | ObjectId | |
+| `name` | string | guest name |
+| `email` | string | for confirmation + reminder |
+| `phone` | string | optional |
+| `partySize` | number | number of covers |
+| `date` | string (`YYYY-MM-DD`) | requested date |
+| `time` | string (`HH:mm`) | requested slot |
+| `tableId` | ObjectId \| null | assigned table (nullable until seated) |
+| `status` | string | `pending` → `confirmed` → `seated` / `cancelled` / `no_show` |
+| `notes` | string | allergies, occasion, seating pref |
+| `depositPaid` | boolean | true after Stripe webhook (if deposits on) |
+| `reminderSentAt` | Date \| null | set by the cron job so it never double-sends |
+| `createdAt` | Date | index + sort |
+
+### `tables` (optional — table booking)
+| field | type | notes |
+|-------|------|-------|
+| `_id` | ObjectId | |
+| `label` | string | "T1", "Window 4" |
+| `seats` | number | capacity |
+| `zone` | string | "patio", "main", "bar" |
+
+Availability = look up reservations for a `date`/`time` window and subtract from
+`tables` capacity. Add a unique index on `{ tableId, date, time }` to prevent
+double-booking a specific table, and index `reservations.createdAt` and
+`{ date: 1, time: 1 }` for the availability query and the reminder job.
+
+---
+
+## 3. How CloudGrid injects things (the whole point)
+
+You never provision infra or set connection strings by hand. Declare intent in
+`cloudgrid.yaml`; the platform provisions and injects env vars.
+
+- **Database (Mongo).** `needs: { database: true }` → the grid provisions shared
+  Mongo and injects the connection string as **`DATABASE_MONGODB_URL`** (plus the
+  legacy **`MONGODB_URL`** alias) at `grid dev` (locally) and at runtime (after
+  `grid plug`). Read it **lazily** inside a getter, `process.env.DATABASE_MONGODB_URL
+  || process.env.MONGODB_URL` — **never at module top level**, or `next build`
+  throws when it imports the module for route analysis before the var exists.
+  `db.client.db()` returns the DB from the connection-string path segment.
+  Do NOT set `DATABASE_MONGODB_URL`/`MONGODB_URL` yourself.
+
+- **Secrets (Stripe, email, auth) → `vault:` block.** Map an org vault item to an
+  env var in `cloudgrid.yaml` `vault:` (already uncommented in this blueprint):
+  `STRIPE_KEY: stripe-live-key`, `SENDGRID_API_KEY: sendgrid-key`,
+  `AUTH_PROVIDER_KEY: auth-provider-secret`. The platform reads the named item
+  from the org vault and injects it as that env var at runtime. Store the actual
+  secret value with `gridctl_secrets` (or the org vault UI) — never commit it. In
+  code read `process.env.STRIPE_KEY`, `process.env.SENDGRID_API_KEY`, etc. Vault
+  is the CloudGrid-correct path for third-party keys; `gridctl_env` is for
+  non-secret config only.
+
+- **AI (optional).** If you add AI (e.g. menu-copy generation, natural-language
+  booking parsing), declare `needs: { ai: true }` → the grid injects
+  **`AI_GATEWAY_URL`**; call it via `@cloudgrid-io/ai`. Not required for the base
+  blueprint.
+
+Reserved env var names you must NOT set: `PORT`, `APP_NAME`, `SERVICE_NAME`,
+`NODE_ENV`, `MONGODB_URL`, `REDIS_URL`, `AI_GATEWAY_URL`, `CLOUDGRID_*`.
+
+---
+
+## 4. Wiring auth and/or payments
+
+### Payments — Stripe deposit to hold a table (optional)
+1. `vault: { STRIPE_KEY: stripe-live-key }` (present in `cloudgrid.yaml`); add
+   `stripe` to `services/web/package.json`.
+2. `app/api/checkout/route.js` — read `process.env.STRIPE_KEY`, construct the
+   Stripe client, create a Checkout Session for the deposit amount, tuck the
+   `reservationId` into the session `metadata`, return the session URL; the client
+   redirects to it.
+3. `app/api/webhooks/stripe/route.js` — verify the Stripe signature, and on
+   `checkout.session.completed` set `depositPaid: true` (and `status: confirmed`)
+   on the reservation from `metadata.reservationId`. This route MUST be public
+   (exclude it from auth middleware) and read the raw request body for signature
+   verification. Register its URL in the Stripe dashboard as the webhook endpoint
+   after the app is live.
+
+### Auth — staff dashboard login (optional)
+1. `vault: { AUTH_PROVIDER_KEY: auth-provider-secret }` (uncomment it), add the
+   provider SDK (e.g. `@clerk/nextjs` or `@auth0/nextjs-auth0`) to
+   `package.json`.
+2. `middleware.js` — protect `/staff/*` (and staff-only API routes) with the
+   provider; leave `/`, `/reserve`, `/api/menu`, `/api/reservations` (POST),
+   `/api/webhooks/stripe` public.
+3. Initialize the SDK from `process.env.AUTH_PROVIDER_KEY` (+ any public
+   publishable key via `gridctl_env`). The public booking flow needs no login.
+
+Keep the public reservation POST open; only staff read/mutate endpoints
+(`GET /api/reservations`, `PATCH/DELETE /api/reservations/[id]`, `/staff/*`) sit
+behind auth.
+
+---
+
+## 5. Deploy steps
+
+Runtime app → **local edition** only (Claude Desktop / Claude Code / CLI); the
+build is **asynchronous**.
+
+1. `gridctl_login_status` (→ `gridctl_login` if needed). Pick the grid if the
+   user has more than one.
+2. `gridctl_init` an app `<name>` — creates the entity + `.cloudgrid/link.json`
+   and a `cloudgrid.yaml` with empty `services: {}`. Run `init` FIRST (plug needs
+   a linked dir).
+3. **Fill** — write the app under `services/web/`, then set `cloudgrid.yaml` to
+   this blueprint's active fields: `name`, `services.web { type: nextjs, path: / }`,
+   `needs: { database: true }`, and the `vault:` mappings you actually use.
+4. Store secrets: `gridctl_secrets` for the vault item values (Stripe/SendGrid/
+   auth). Non-secret config (publishable keys, feature flags) → `gridctl_env`.
+5. (Optional) `grid dev` — runs Next.js locally with `DATABASE_MONGODB_URL` and
+   vault vars injected against dev Mongo. Seed `menu`, test a booking.
+6. `gridctl_plug` — builds + deploys `services/web/`. **Async**: the first
+   response is `status: building`, not a URL. Poll `gridctl_status` (or the
+   returned poll URL) until live; surface a liveness signal while it builds
+   (never a bare silent wait). Only then return the **live app URL** (not the
+   build/log link).
+7. Iterate by re-plugging the SAME entity — it updates the same URL. If you
+   turned on Stripe, register the live `/api/webhooks/stripe` URL in Stripe.
+
+---
+
+## 6. Edition note
+
+This is a **runtime app** (built + deployed container), so it requires the
+**local edition** (Claude Desktop / Claude Code) or the CLI. On the **hosted**
+edition (Claude Web / hosted MCP) you CANNOT build a runtime app — hosted is
+inline-only and can publish static pages only. If the user is on hosted, say so
+plainly and offer a static, non-persistent menu page instead (no bookings), then
+stop the runtime path.
+
+---
+
+## 7. Cron reminders — INTENDED SHAPE, NOT YET DEPLOYABLE (#1543)
+
+The design calls for a scheduled job that emails confirmation reminders to guests
+with a booking the next day. On CloudGrid this is a second **cron service** in the
+same entity:
+
+```yaml
+services:
+  reminder:
+    type: cron
+    schedule: "0 15 * * *"   # daily 15:00
+    timezone: UTC
+    run: job                 # runs services/reminder/ as a one-shot job
+```
+
+The cron job (`services/reminder/index.js`) reads the SAME grid-injected
+`DATABASE_MONGODB_URL` and the vault-injected `SENDGRID_API_KEY`, queries
+`reservations` where `date` is tomorrow and `reminderSentAt` is null and
+`status` is `confirmed`, sends each a reminder, and stamps `reminderSentAt` so it
+never double-sends.
+
+> **PENDING platform issue #1543:** cron deploy is not yet available. Keep the
+> `reminder:` cron block **commented out** in `cloudgrid.yaml` (as shipped) —
+> plugging an entity with an active cron service will be rejected until #1543
+> lands. Build the rest now; enable the cron block once cron ships. Interim
+> workaround: trigger the same reminder logic from an external scheduler hitting
+> a protected HTTP route, or run it manually.
+
+---
+
+## 8. DB wiring reference (copy the app-with-data shape)
+
+`services/web/lib/db.js` — lazy cached Mongo client (identical pattern to the
+`app-with-data` template):
+
+```js
+import { MongoClient } from "mongodb";
+
+function clientPromise() {
+  const uri = process.env.DATABASE_MONGODB_URL || process.env.MONGODB_URL;
+  if (!uri) throw new Error("DATABASE_MONGODB_URL not set — the grid injects it; use grid dev / grid plug.");
+  if (!globalThis.__mongoClientPromise) {
+    globalThis.__mongoClientPromise = new MongoClient(uri).connect();
+  }
+  return globalThis.__mongoClientPromise;
+}
+
+export async function getDb() {
+  const client = await clientPromise();
+  return client.db();
+}
+```
+
+Every route/server component imports `getDb()` from here. Mark reservation and
+menu routes `export const dynamic = "force-dynamic"` so they always hit Mongo.
+Validate `partySize`, `date`, `time`, `email` on the server before insert.

package/src/corpus/templates/restaurant-reservations/README.md

@@ -0,0 +1,17 @@
+# restaurant-reservations template — BLUEPRINT (structure + cloudgrid.yaml)
+
+A restaurant website with online reservations: a public Next.js site (menu
+content + a booking form / table booking) backed by grid-shared MongoDB, with
+CloudGrid-native wiring for the database, third-party secrets (Stripe deposit,
+email/SMS via the `vault:` block), optional staff auth, and a scheduled reminder
+job. Reservations persist in Mongo (`needs: { database: true }` → injected
+`DATABASE_MONGODB_URL`); this is a runtime app (local edition, async deploy) —
+not a static page.
+
+**This is a blueprint: structure + `cloudgrid.yaml`, adapt and build.** There is
+no app code here. Read `AGENTS.md` for the file tree, the Mongo collections
+(`menu`, `reservations`, `tables`), how CloudGrid injects the DB and vault
+secrets, how to wire Stripe/auth, and the deploy steps — then build the app under
+`services/web/` following it. Note: the confirmation-reminder **cron** service is
+documented but NOT yet deployable (platform issue #1543) — keep that block
+commented.

package/src/corpus/templates/restaurant-reservations/cloudgrid.yaml

@@ -0,0 +1,258 @@
+# ============================================================================
+# cloudgrid.yaml — Complete Example
+# ============================================================================
+#
+# This file shows EVERY field the platform supports, with comments explaining
+# each one. It is a reference — no real project would use all of these at once.
+#
+# STANDARD: every CloudGrid template ships its cloudgrid.yaml in THIS full form.
+# Uncomment ONLY the fields the archetype needs (always: name + services; plus
+# needs: for runtime apps). Leave everything else commented. Comments are
+# ignored by the parser, so the file deploys to exactly the active fields.
+#
+# Only two fields are required: name and services. Everything else is optional.
+# Full docs: docs/cloudgrid-yaml-reference.md
+# ============================================================================
+
+
+# ─── IDENTITY ───────────────────────────────────────────────────────────────
+
+# Entity slug (REQUIRED). 2-42 chars, lowercase alphanumeric + hyphens.
+# Must start and end with a letter or digit.
+# Becomes part of your public URL: <name>-<4hex>.<org>.cloudgrid.io
+name: restaurant-reservations
+
+# One-line summary. Max 280 chars. If omitted, Grid Brain auto-generates one.
+# description: Order processing API for the e-commerce platform
+
+# Single emoji or short text shown as the entity icon. Max 64 chars.
+# icon: "📦"
+
+# Entity type: 'app' (default) or 'agent'. An agent is an app with an agent: block.
+# kind: app
+
+
+# ─── NETWORKING ─────────────────────────────────────────────────────────────
+
+# Whether this entity serves public HTTP traffic.
+# true = public URL + Ingress (default). false = internal only.
+# expose: true
+
+# Additional domain names this entity serves on (the platform URL is always active).
+# custom_domains:
+#   - orders.example.com
+#   - shop.example.com
+
+
+# ─── HEALTH PROBE (entity-level override) ──────────────────────────────────
+
+# Override the default health check for the entire entity.
+# probe:
+#   path: /healthz               # HTTP path to probe (default: /).
+#   expected: "<500"             # '<500' (default), '200', '2xx'.
+
+
+# ─── DEPLOY NOTIFICATIONS ──────────────────────────────────────────────────
+
+# Slack notification to the org's default channel on deploy failure.
+# true = notify (default), false = silent.
+# notify: true
+
+
+# ─── SCALING ────────────────────────────────────────────────────────────────
+
+# Replica scaling intent.
+#   'auto'  — platform decides from the grid tier (default).
+#   'alert' — single-replica, emit a SPOF warning at plug time.
+#   1-10    — explicit replica count, clamped to the grid max.
+# scale: auto
+
+
+# ─── INFRASTRUCTURE NEEDS ──────────────────────────────────────────────────
+#
+# Declare infrastructure; the platform provisions it and injects connection
+# details as env vars. Nine needs: database, cache, kv, queue, pubsub, vector,
+# object_storage, disk, ai. Simplest form is `true`.
+#
+# needs:
+#   # database — MongoDB. Injects DATABASE_MONGODB_URL (+ legacy MONGODB_URL).
+#   #   true | pool | dedicated | { tier: external, secret: MY_DB }
+#   database: true
+#   # cache — Redis, LRU eviction (derived data). Injects CACHE_REDIS_URL (+ REDIS_URL).
+#   #   true | redis | { engine: external, secret: MY_REDIS_URL }
+#   cache: true
+#   # kv — Redis, no eviction (durable: flags, idempotency). Injects KV_REDIS_URL.
+#   kv: true
+#   # queue — Redis job-queue substrate (BullMQ). Injects QUEUE_REDIS_URL.
+#   queue: true
+#   # pubsub — Redis broadcast pub/sub. Injects PUBSUB_REDIS_URL.
+#   pubsub: true
+#   # vector — pgvector on Postgres. Injects VECTOR_PGVECTOR_URL (+ PGVECTOR_URL).
+#   #   true | pgvector | { engine: pgvector, dim: 1536 }
+#   vector: pgvector
+#   # object_storage — GCS bucket. Injects OBJECT_STORAGE_GCS_BUCKET, _GCS_REGION.
+#   #   true | { size: 10Gi }
+#   object_storage: true
+#   # disk — persistent FS mount (PVC). Injects DISK_PATH. Pins to 1 replica.
+#   #   true | { size: 1Gi, mount: /var/data }  (forbidden: /proc /sys /dev)
+#   disk: true
+#   # ai — AI Gateway via @cloudgrid-io/ai. Injects AI_GATEWAY_URL. true | false.
+#   ai: true
+
+
+# ─── INTER-ENTITY COMMUNICATION ────────────────────────────────────────────
+
+# Entities this app calls (outbound). Service-account token injected per pair.
+# calls:
+#   - org:atomic/billing-api
+
+# Inbound caller policy — who can call YOUR entity.
+# callers:
+#   policy: auto                 # 'auto' | 'manual' (default).
+#   scope: org                   # 'org' (default) | 'space' | 'explicit'.
+#   fallback: deny               # 'deny' (default) | 'warn'.
+#   # allowed_apps:              # required only when scope: explicit
+#   #   - org:atomic/frontend
+
+
+# ─── VAULT ──────────────────────────────────────────────────────────────────
+#
+# Map org vault items to env vars. Keys UPPER_SNAKE_CASE; values = vault item keys.
+# This is the CloudGrid-correct way to inject third-party secrets (auth provider,
+# Stripe, email). The platform reads the named item from the org vault and injects
+# it as the env var at runtime — never hardcode or commit these.
+#
+# This blueprint uses a booking-provider / notification secret and (optional)
+# Stripe deposit for reservations. Uncomment only the ones you actually wire.
+vault:
+  # Stripe secret key — required only if you take a card deposit to hold a table.
+  # Used by the Stripe SDK in the deposit checkout + webhook routes.
+  STRIPE_KEY: stripe-live-key
+  # Transactional email/SMS provider key for confirmation + reminder messages.
+  SENDGRID_API_KEY: sendgrid-key
+  # Auth provider (Clerk/Auth0) key — only if the restaurant staff dashboard
+  # needs a login. Public reservation form needs no auth.
+  # AUTH_PROVIDER_KEY: auth-provider-secret
+
+
+# ─── CONNECTORS AND HOOKS ──────────────────────────────────────────────────
+
+# connectors:
+#   - id: team-slack
+#     target: org                # 'system' | 'org'.
+#     channel: "#deploys"
+# hooks:
+#   - on: deploy.failed          # release.published | deploy.failed | feedback.received | entity.created
+#     via: team-slack
+#     template: "Deploy failed for {entity_name}: {error}"
+#     channel: "#alerts"
+# triggers:
+#   - from: team-slack
+#     event: message
+#     channel: "#support"
+#     route: /webhooks/slack
+
+
+# ─── LOCAL DEVELOPMENT ──────────────────────────────────────────────────────
+
+# Config for `grid dev`. Redact fields in dev data snapshots.
+# dev:
+#   snapshot:
+#     mask:
+#       - reservations.email
+#       - reservations.phone
+
+
+# ─── AGENT METADATA ────────────────────────────────────────────────────────
+
+# An agent is an app with kind: agent (above) and an agent: block.
+# agent:
+#   purpose: Sync customer data from external CRM every night
+#   schedule: "0 2 * * *"
+#   trigger: webhook
+
+
+# ─── PRIVATE RESOURCE AUTO-DEPENDENCIES ────────────────────────────────────
+
+# When true (default), private resources are auto-injected as depends_on.
+# auto_depends_on_private_requires: true
+
+
+# ─── SERVICES (REQUIRED) ───────────────────────────────────────────────────
+#
+# At least one service is required. Key = service name (slug rules).
+# Types: node, nextjs, python, static, cron.
+#
+services:
+  web:
+    # type (REQUIRED): node | nextjs | python | static | cron
+    type: nextjs
+    # HTTP route: '/' | '/api' | false (internal) | omitted (=false + warning).
+    # Only one service per entity can claim '/'.
+    path: /
+    # Container port (default 8080; read process.env.PORT). Not for cron.
+    # port: 3000
+    # nextjs only: base_image: distroless (default) | alpine
+    # base_image: distroless
+    # node only: lang: javascript (default) | typescript
+    # lang: typescript
+    # Custom source location (default services/<name>/). '.' = project root.
+    # source:
+    #   path: src/worker
+    # Health probe: false/omitted (TCP) | true (/health) | "/path" | {path,expected}
+    # health:
+    #   path: /api/health
+    #   expected: "200"
+    # Static runtime env vars (strings). Reserved: PORT, APP_NAME, SERVICE_NAME,
+    # NODE_ENV, MONGODB_URL, REDIS_URL, AI_GATEWAY_URL, N8N_WEBHOOK_URL, CLOUDGRID_*.
+    # env:
+    #   LOG_LEVEL: info
+    # Sibling services/private resources that must be Ready first (no cron, no cycles).
+    # depends_on:
+    #   - worker
+    # Per-service needs override (merges with app-level needs).
+    # needs:
+    #   queue: true
+    # Static-only build step:
+    # build:
+    #   command: npm run build
+    #   output: dist
+    #   env:
+    #     VITE_API_URL: https://api.example.com
+    # node_version: "22"          # static build step: '18'|'20'|'22'|'24' (quoted)
+    # Cron-only: schedule (REQUIRED), timezone (UTC|EST|PST), run (job | https URL)
+    # schedule: "0 9 * * 1"
+    # timezone: UTC
+    # run: job
+    # Verbose per-service persistence (prefer needs.disk):
+    # persist:
+    #   - name: scratch
+    #     path: /tmp/scratch
+    #     size: 500Mi
+
+  # ─── CRON SERVICE (INTENDED SHAPE — NOT YET DEPLOYABLE) ──────────────────
+  # Scheduled confirmation / reminder sender for upcoming reservations.
+  # A cron service has no HTTP path; it runs `run: job` on `schedule`.
+  # NOTE: cron deploy is currently a PENDING platform issue (#1543). This block
+  # documents the intended shape only — leave it COMMENTED until cron ships,
+  # or the plug will reject the entity. See AGENTS.md → "Cron reminders".
+  # reminder:
+  #   type: cron
+  #   schedule: "0 15 * * *"      # daily 15:00 — remind next-day bookings
+  #   timezone: UTC
+  #   run: job                    # runs services/reminder/ as a one-shot job
+  #   # The cron job reads the SAME injected DATABASE_MONGODB_URL + vault secrets
+  #   # (SENDGRID_API_KEY) to find tomorrow's reservations and send reminders.
+
+
+needs:
+  database: true
+
+
+# ============================================================================
+# NOTES
+# - NOT here: org:/entity_id: (→ .cloudgrid/link.json), secrets (→ grid secrets set),
+#   Dockerfile (platform generates from services.*.type).
+# - Deploy: grid plug   |   Local with grid resources: grid dev
+# - Full reference: docs/cloudgrid-yaml-reference.md  |  CLI: docs/cli-reference.md
+# ============================================================================

package/src/corpus/templates/restaurant-reservations/index.md

@@ -0,0 +1,62 @@
+# Template: restaurant-reservations (BLUEPRINT — Next.js + Mongo)
+
+A restaurant website with online reservations. A public Next.js site (menu +
+booking form / table booking) backed by grid-shared MongoDB, with CloudGrid-native
+wiring for the DB, third-party secrets (`vault:`), optional Stripe deposit and
+staff auth, and a scheduled reminder job. Reservations persist in Mongo — this is
+a **runtime app** (local edition, async deploy), not a static page.
+
+**This is a blueprint, not runnable code.** It ships a `cloudgrid.yaml` (active:
+`name` + `services.web` nextjs + `needs: { database: true }` + a `vault:` block)
+and an `AGENTS.md` structure guide. Fetch both, read `AGENTS.md`, then build the
+app under `services/web/` following it — the proven `app-with-data` shape with a
+`reservations` domain.
+
+**Key rules:**
+
+1. **App code MUST live under `services/web/`**, not the template root. `path: /`
+   is the URL mount, NOT the filesystem path.
+2. **Read `DATABASE_MONGODB_URL` LAZILY** (inside the getter), legacy
+   `MONGODB_URL` fallback — the grid injects it; never hardcode, never commit.
+3. **Third-party secrets go through the `vault:` block** (Stripe `STRIPE_KEY`,
+   email `SENDGRID_API_KEY`, optional `AUTH_PROVIDER_KEY`) → injected env vars;
+   store values with `gridctl_secrets`.
+4. **Declare `needs: { database: true }`** (canonical) — never mix `needs:` and
+   `requires:`.
+5. **Cron reminder service is documented but NOT deployable yet** (platform issue
+   #1543) — keep the `reminder:` cron block commented.
+
+## Fetch bundle
+
+```
+gridctl_fetch("template", "restaurant-reservations")   # cloudgrid.yaml + AGENTS.md + README
+```
+
+Then read `AGENTS.md` (file tree, `menu`/`reservations`/`tables` collections,
+CloudGrid env/vault/deploy wiring, Stripe + auth, cron shape) and build under
+`services/web/`. Deploy: `grid init` → fill → `grid plug` (async — poll to a live
+URL). Local edition only.
+
+## cloudgrid.yaml (active fields)
+
+```yaml
+# On disk this file is the full-annotated reference (templates/_cloudgrid.yaml.reference)
+# with EVERY field present as a comment; only the fields below are uncommented
+# (plus a vault: block and a COMMENTED cron reminder service), so it deploys to
+# exactly these active fields.
+name: restaurant-reservations
+vault:
+  STRIPE_KEY: stripe-live-key
+  SENDGRID_API_KEY: sendgrid-key
+services:
+  web:
+    type: nextjs
+    path: /
+needs:
+  database: true
+```
+
+> **Capability:** `needs: { database: true }` → the deployer provisions Mongo and
+> injects `DATABASE_MONGODB_URL` (+ legacy `MONGODB_URL`). Vault items are injected
+> as the mapped env vars. `requires:` is the deprecated v1 alias — don't mix it
+> with `needs:`. See the capability-map for the full injection table.

package/src/corpus/templates/revenue-dashboard/README.md

@@ -0,0 +1,75 @@
+# revenue-dashboard template — persistent Next.js + Mongo revenue dashboard
+
+A minimal but real, deployable revenue/sales dashboard. Sales live in the
+grid-shared MongoDB, so they survive refresh and are shared across
+users/sessions — unlike a static page whose state is in memory only.
+
+It shows a **total revenue** figure, a **per-product breakdown**, a **recent
+sales table**, and a form to **add a sale**. Each sale is a document in a `sales`
+collection: `{ product: string, amount: number, at: Date }`.
+
+## How the grid gives you a database
+
+You do **not** provision a database or set a connection string. In
+`cloudgrid.yaml` you declare `needs: { database: true }`, and the grid:
+
+- provisions shared Mongo for the app, and
+- injects the connection string as the **`DATABASE_MONGODB_URL`** environment
+  variable (plus the legacy `MONGODB_URL` alias) — at dev-time (`grid dev`) and
+  at runtime (after `grid plug`).
+
+The app reads it via `process.env.DATABASE_MONGODB_URL` (with a legacy
+`process.env.MONGODB_URL` fallback) in `services/web/lib/db.js` — **lazily,
+inside the `getDb` getter, never at module top level** (a top-level read fails
+`next build`, which imports the module for route analysis before the grid injects
+the var). Never hardcode a connection string; never commit a secret.
+
+> **Declare `needs: { database: true }` (the canonical shape).** The deployer
+> provisions Mongo and injects `DATABASE_MONGODB_URL` (plus the legacy
+> `MONGODB_URL` alias). `requires:` is the deprecated v1 alias — don't author new
+> yaml with it, and never set `needs:` and `requires:` together (the validator
+> rejects the combination).
+
+## Service layout
+
+App code lives under **`services/web/`**, not the template root. `path:` in
+`cloudgrid.yaml` is the URL mount, not the filesystem path — the service named
+`web` means the CLI looks for `services/web/`. Files at the root fail with
+`Error: Service directory not found: …/services/web`.
+
+## Run locally
+
+```bash
+npm install
+grid dev          # runs Next.js with DATABASE_MONGODB_URL injected against dev Mongo
+```
+
+## Deploy
+
+```bash
+grid plug         # builds + deploys the folder (async — poll status until live)
+```
+
+A runtime deploy is asynchronous: `plug` returns `status: building`; poll status
+until the app is live, then use the returned live URL. Re-plug the same entity
+to update the same URL.
+
+## File tree
+
+```
+cloudgrid.yaml                        # name + services.web (nextjs) + needs: { database: true }
+services/web/package.json             # next, react, react-dom, mongodb driver only
+services/web/lib/db.js                # lazy Mongo client from DATABASE_MONGODB_URL (legacy MONGODB_URL fallback)
+services/web/app/layout.js            # root layout + inline CSS
+services/web/app/page.js              # server component: total + per-product breakdown + recent table
+services/web/app/sale-form.js         # client form: POST a sale, refresh the server component
+services/web/app/api/sales/route.js   # GET (list) / POST (add) / DELETE (remove)
+```
+
+## Adapt it
+
+- Rename the `sales` collection in `services/web/app/api/sales/route.js` and
+  `services/web/app/page.js` (e.g. `invoices`, `orders`, `payments`).
+- Change the document fields (add currency, region, customer, MRR flags).
+- Add a chart, date filters, or per-period rollups as the app grows.
+- Add `cache: true` to `needs:` only if you actually need Redis.