Fase 6: deploy readiness (PWA + invitation rate-limit + JWT rotation)

Closes the deferrals from Fase 4 that gated a public deploy of the MVP.

PWA install
- Custom SW at apps/web/src/sw.ts (named sw.ts, not service-worker.ts, so
  SvelteKit doesn't intercept it and block Workbox imports).
- vite.config.ts switched to injectManifest strategy with precache of the
  built shell; no runtime caching of Supabase (offline writes stay on the
  existing $lib/sync pending_ops queue).
- Root +layout.svelte onMount() calls registerSW({ immediate: true }) from
  virtual:pwa-register — the plugin does not auto-register.
- Web manifest with 192/512/512-maskable icons + iOS meta tags
  (apple-mobile-web-app-capable, apple-touch-icon, etc.) in app.html.
- Placeholder icons generated via ImageMagick; replace before public launch
  (noted in apps/web/static/icons/README.md).

Kong rate-limit
- /rest/v1/collective_invitations POST: 20/hour per Authorization header.
  Anti-spam on invitation creation is the only rate-limit that survived —
  auth rate-limits were dropped during execution (see plan §6.3): Keycloak
  already provides bruteForceProtected=true on the realm, and stacking a
  Kong limit on /auth/v1/token throttled legitimate PKCE code exchanges
  during E2E.
- Added 'rate-limiting' to KONG_PLUGINS in docker-compose.dev.yml.
- Discovered: strip_path=true on a full-table path sends "/" upstream and
  PostgREST rejects POST to root with PGRST117. Route carries a
  request-transformer plugin that rewrites the URI back.

JWT rotation + prod template
- infra/scripts/rotate-jwt.sh signs anon + service_role JWTs with a fresh
  48-byte secret via openssl. Prints three values for the operator to
  paste; does not touch files.
- Dev secret rotated as a dry-run. Updated both .env (local, gitignored)
  and apps/web/.env.development. Existing sessions are invalidated — all
  users must re-login once.
- .env.production.example committed with placeholders + rotation notes.

Lighthouse + RLS audit tooling
- `just lighthouse` boots the prod build + Supabase stack and runs
  lighthouse CLI against PWA/A11y/Best-Practices.
- `just rls-audit` runs infra/scripts/rls-audit.sh which signs an Eva JWT
  (non-member) and GETs 8 REST endpoints — all must return []. Complements
  the Vitest rls-audit.test.ts suite.

Tests
- 236 green: 34 pgTAP + 140 Vitest integration + 15 unit + 46 Playwright.
- 1 rate-limit test gated behind RUN_RATE_LIMIT_TESTS=1 (Kong counters are
  shared state); run via `just test-rate-limit` which force-recreates Kong
  first to reset counters.
- 3 skipped in `just test-all` (2 Realtime presence upstream bug, 1 gated
  rate-limit).

Deliberately out of scope: push notifications, CI ephemeral Docker, final
production icons, runtime caching of Supabase (SyncBanner + pending_ops
already covers offline).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
2026-04-14 04:48:04 +02:00
parent 1f5b34d55d
commit 2db5aaac14
24 changed files with 571 additions and 55 deletions

View File

@@ -0,0 +1,84 @@
/**
* RL-series: Kong rate-limiting on the invitation endpoint (Fase 6.3)
*
* Kong ships with the rate-limiting plugin; we configure `local` policy
* (single-instance). This test hits the Supabase Kong gateway on localhost
* and asserts that the 21st POST in the window returns 429.
*
* Scope note: we deliberately do NOT rate-limit /auth/v1/* at the Kong layer.
* Keycloak's native `bruteForceProtected` flag on the `colectivo` realm
* (maxFailureWaitSeconds=900, failureFactor=30) covers password attacks at the
* IdP level — the authoritative place. A Kong limit on /auth/v1/token would
* also throttle legitimate PKCE code exchanges during tests.
*
* Threshold (in sync with infra/kong.yml):
* /rest/v1/collective_invitations POST — 20 per hour per authenticated user
*/
import { describe, it, expect } from 'vitest';
import { createClientAs } from '../src/supabase-clients.js';
import { ANA_ID, COLLECTIVE_ID } from '../src/seed-constants.js';
const KONG = process.env.PUBLIC_SUPABASE_URL ?? 'http://localhost:8001';
const ANON = process.env.PUBLIC_SUPABASE_ANON_KEY!;
// These tests hit Kong's real rate-limit counters (IP + token-based) and
// consume quota that's shared with the rest of the integration suite. To
// avoid cross-test pollution in `just test-all`, they're gated behind an
// explicit env flag and run via `just test-rate-limit` (which also
// restarts Kong first to ensure a clean counter state).
const RUN = process.env.RUN_RATE_LIMIT_TESTS === '1';
const d = RUN ? describe : describe.skip;
d('Kong rate-limit — /rest/v1/collective_invitations', () => {
it('RL-02: 21st invitation POST in an hour returns 429', async () => {
const ana = await createClientAs(ANA_ID);
const authHeader = (ana as unknown as { rest: { headers: Record<string, string> } }).rest
?.headers?.Authorization;
// Fallback: grab the token manually
const token = authHeader?.replace('Bearer ', '') ?? (await signBorja());
const headers = {
'Content-Type': 'application/json',
apikey: ANON,
Authorization: `Bearer ${token}`,
Prefer: 'return=representation'
};
for (let i = 0; i < 20; i++) {
const r = await fetch(`${KONG}/rest/v1/collective_invitations`, {
method: 'POST',
headers,
body: JSON.stringify({
collective_id: COLLECTIVE_ID,
role: 'member',
created_by: ANA_ID
})
});
expect(r.status, `attempt ${i + 1} was 429`).not.toBe(429);
}
const limited = await fetch(`${KONG}/rest/v1/collective_invitations`, {
method: 'POST',
headers,
body: JSON.stringify({
collective_id: COLLECTIVE_ID,
role: 'member',
created_by: ANA_ID
})
});
expect(limited.status).toBe(429);
}, 60_000);
});
// Helper — sign a minimal authenticated JWT for Ana if we need to fall back.
async function signBorja(): Promise<string> {
const { SignJWT } = await import('jose');
const secret = new TextEncoder().encode(process.env.SUPABASE_JWT_SECRET!);
return new SignJWT({ role: 'authenticated' })
.setProtectedHeader({ alg: 'HS256' })
.setSubject(ANA_ID)
.setAudience('authenticated')
.setIssuedAt()
.setExpirationTime('1h')
.sign(secret);
}