EventSnap/e2e/specs/07-adversarial/auth-tampering.spec.ts

/**
 * Phase 2 adversarial — JWT forgery, brute-force, and password attacks.
 *
 * The JWT secret is in docker-compose.test.yml as a fixed value — these
 * tests do NOT try to forge tokens using that secret (that would only
 * prove HS256 works). Instead they assert the *failure* paths: alg:none,
 * tampered signature, expired sessions, wrong role.
 */
import { test, expect } from '../../fixtures/test';
import { ADMIN_PASSWORD } from '../../fixtures/api-client';
import { BASE } from '../../helpers/env';

/** RFC-4648 base64url with no padding. */
function b64u(s: string) {
  return Buffer.from(s)
    .toString('base64')
    .replace(/=+$/, '')
    .replace(/\+/g, '-')
    .replace(/\//g, '_');
}

test.describe('Adversarial — JWT', () => {
  test('alg:none token claiming admin role is rejected', async () => {
    const header = b64u(JSON.stringify({ alg: 'none', typ: 'JWT' }));
    const payload = b64u(
      JSON.stringify({
        sub: '00000000-0000-0000-0000-000000000000',
        role: 'admin',
        event_id: '00000000-0000-0000-0000-000000000000',
        exp: Math.floor(Date.now() / 1000) + 3600,
      })
    );
    const token = `${header}.${payload}.`;
    const res = await fetch(`${BASE}/api/v1/admin/config`, {
      headers: { Authorization: `Bearer ${token}` },
    });
    expect(res.status).toBe(401);
  });

  test('JWT with valid structure but bogus signature is rejected', async ({ guest }) => {
    const g = await guest('SigForge');
    const parts = g.jwt.split('.');
    // Replace the signature with random bytes of the same length.
    const fakeSig = parts[2].split('').reverse().join('');
    const tampered = `${parts[0]}.${parts[1]}.${fakeSig}`;
    const res = await fetch(`${BASE}/api/v1/me/context`, {
      headers: { Authorization: `Bearer ${tampered}` },
    });
    expect(res.status).toBe(401);
  });

  test('JWT with payload-tampered role=admin (re-encoded payload, original signature) is rejected', async ({
    guest,
  }) => {
    const g = await guest('RolePromote');
    const parts = g.jwt.split('.');
    const original = JSON.parse(Buffer.from(parts[1], 'base64url').toString());
    const escalated = { ...original, role: 'admin' };
    const newPayload = b64u(JSON.stringify(escalated));
    const tampered = `${parts[0]}.${newPayload}.${parts[2]}`;
    const res = await fetch(`${BASE}/api/v1/admin/config`, {
      headers: { Authorization: `Bearer ${tampered}` },
    });
    // Signature won't match the new payload → middleware must return 401, not 403.
    expect(res.status).toBe(401);
  });

  test('JWT for a session that was deleted (logout) is rejected', async ({ guest, api }) => {
    const g = await guest('LoggedOut');
    await api.logout(g.jwt);
    const res = await fetch(`${BASE}/api/v1/me/context`, {
      headers: { Authorization: `Bearer ${g.jwt}` },
    });
    expect(res.status).toBe(401);
  });

  test('Authorization header without "Bearer " prefix is rejected', async ({ guest }) => {
    const g = await guest('NoBearer');
    const res = await fetch(`${BASE}/api/v1/me/context`, {
      headers: { Authorization: g.jwt },
    });
    expect([401, 403]).toContain(res.status);
  });

  test('missing Authorization header on protected route returns 401', async () => {
    const res = await fetch(`${BASE}/api/v1/me/context`);
    expect(res.status).toBe(401);
  });
});

test.describe('Adversarial — PIN brute-force', () => {
  /**
   * The PIN defence has two tiers, and telling them apart is the whole point of these tests:
   *
   *   - the per-(IP, name) throttle, which refuses the ATTACKER; and
   *   - the account lock, which refuses the VICTIM — the only tier that can be weaponised.
   *
   * Both answer 429, so the status code alone proves nothing. The regression these guard is that
   * the lock threshold used to sit BELOW the throttle ceiling (3 vs 5), so three requests from a
   * single IP locked any guest whose display name is readable off the feed, every 15 minutes,
   * indefinitely. The tier meant to protect a guest was the cheapest way to attack them.
   */
  // Restore the default. The first test turns the limiter on for the whole instance, and leaving
  // it on would throttle unrelated specs sharing this stack. Runs even on failure.
  test.afterEach(async ({ api, adminToken }) => {
    await api.patchConfig(adminToken, { rate_limits_enabled: 'false' });
  });

  test('a single IP is throttled without ever locking the victim out', async ({
    api,
    adminToken,
    guest,
    db,
  }) => {
    // Rate limits are off by default in this environment, and the throttle IS the tier under
    // test — without it the run would silently assert only half the property.
    await api.patchConfig(adminToken, {
      rate_limits_enabled: 'true',
      recover_rate_enabled: 'true',
    });

    const g = await guest('Brute');
    const wrong = g.pin === '0000' ? '1111' : '0000';

    // Serially, so the failed-PIN counter increments monotonically. Well past the per-(IP, name)
    // ceiling of 4, and past the OLD lock threshold of 3.
    const statuses: number[] = [];
    for (let i = 0; i < 8; i++) {
      const r = await fetch(`${BASE}/api/v1/recover`, {
        method: 'POST',
        headers: { 'Content-Type': 'application/json' },
        body: JSON.stringify({ display_name: g.displayName, pin: wrong }),
      });
      statuses.push(r.status);
    }
    expect(statuses.filter((s) => s === 200), 'a wrong PIN must never authenticate').toHaveLength(
      0
    );
    expect(statuses.some((s) => s === 429), 'the attacker must be throttled').toBe(true);

    expect(
      await db.isPinLocked(g.userId),
      'one IP must not be able to lock a guest out of their own account'
    ).toBe(false);
  });

  test('the account still locks once the failure count is reached', async ({ guest, db }) => {
    const g = await guest('BruteDistributed');
    const wrong = g.pin === '0000' ? '1111' : '0000';

    // The per-IP throttle is what a single source hits first, so drive the counter the way a
    // DISTRIBUTED attacker would — the tier this test covers is the last line against exactly
    // that, and it must not have been removed while fixing the weaponisation above.
    await db.setFailedPinAttempts(g.userId, 11);

    const r = await fetch(`${BASE}/api/v1/recover`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json', 'X-Forwarded-For': '203.0.113.77' },
      body: JSON.stringify({ display_name: g.displayName, pin: wrong }),
    });
    expect(r.status).toBe(401);

    expect(
      await db.isPinLocked(g.userId),
      'a distributed guesser must still trip the account lock'
    ).toBe(true);

    // And the lock holds even against the correct PIN, which is what makes it a real control.
    const correct = await fetch(`${BASE}/api/v1/recover`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json', 'X-Forwarded-For': '203.0.113.78' },
      body: JSON.stringify({ display_name: g.displayName, pin: g.pin }),
    });
    expect(correct.status).toBe(429);
  });

  test('the wrong-PIN streak is atomic under concurrency', async ({ guest, db }) => {
    const g = await guest('BruteParallel');
    const wrong = g.pin === '0000' ? '1111' : '0000';

    await Promise.all(
      Array.from({ length: 10 }, () =>
        fetch(`${BASE}/api/v1/recover`, {
          method: 'POST',
          headers: { 'Content-Type': 'application/json' },
          body: JSON.stringify({ display_name: g.displayName, pin: wrong }),
        })
      )
    );

    // How many of the 10 get past the throttle is genuinely racy and cannot be asserted. What is
    // NOT racy is that every one that DID reach the handler incremented the counter — it is a
    // single `SET x = x + 1 ... RETURNING`, so none of them can be lost to the race.
    expect(
      await db.failedPinAttempts(g.userId),
      'concurrent wrong PINs must all be counted'
    ).toBeGreaterThan(1);
  });
});

test.describe('Adversarial — admin password brute-force', () => {
  // These tests deliberately exhaust the admin-login limiter for the shared test IP. That creates a
  // chicken-and-egg for the NEXT test: its truncate auto-fixture must itself call admin_login before
  // it can reset the counter — so a still-full window would lock the fixture out with 429 before it
  // could clear anything. Disable the toggle here via patchConfig (which authenticates with a JWT,
  // NOT the rate-limited admin_login path), so the next login is clear; truncate then wipes the map.
  // Runs even if the test body failed, so a failure can't poison the rest of the run.
  test.afterEach(async ({ api, adminToken }) => {
    await api.patchConfig(adminToken, { admin_login_rate_enabled: 'false' });
  });

  const tryLogin = (password: string) =>
    fetch(`${BASE}/api/v1/admin/login`, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify({ password }),
    });

  test('admin login is IP rate-limited: a burst of wrong passwords starts returning 429', async ({
    api,
    adminToken,
  }) => {
    // `admin_login` HAS a 5/min/IP throttle (auth/handlers.rs), gated by the master
    // `rate_limits_enabled` + `admin_login_rate_enabled`. Both default TRUE in production — but the
    // e2e reseed turns the master switch OFF for every test, so this defence ran in ZERO tests.
    // (The previous version of this test observed all-401 under that disabled switch and wrongly
    // "documented" that no rate limit exists.) Turn it on and prove it.
    //
    // patchConfig uses the already-minted adminToken (a JWT), so enabling the limiter does not lock
    // us out of configuring it.
    await api.patchConfig(adminToken, {
      rate_limits_enabled: 'true',
      admin_login_rate_enabled: 'true',
    });

    // Sequential, not parallel: the counter increments deterministically so the 429 is not itself
    // subject to the counter race the PIN test covers.
    const statuses: number[] = [];
    for (let i = 0; i < 10; i++) {
      statuses.push((await tryLogin('wrong-' + i)).status);
      if (statuses[i] === 429) break;
    }

    // The password path actually ran (budget existed) before the limiter engaged...
    expect(
      statuses[0],
      'first attempt should be a normal wrong-password 401, not a spurious 429'
    ).toBe(401);
    // ...and the limiter DID engage within the window. Delete the throttle and this is never true.
    expect(
      statuses.some((s) => s === 429),
      'a burst of wrong admin passwords from one IP must start being rate-limited (429)'
    ).toBe(true);
    // No wrong password ever authenticated.
    expect(statuses.some((s) => s === 200)).toBe(false);
  });

  test('once throttled, even the CORRECT admin password is refused (it is an IP limit, not a password check)', async ({
    api,
    adminToken,
  }) => {
    // This is the assertion that makes the test non-vacuous: it isolates the RATE LIMIT from the
    // password logic. If the throttle were removed, the correct password would return 200 here.
    await api.patchConfig(adminToken, {
      rate_limits_enabled: 'true',
      admin_login_rate_enabled: 'true',
    });

    // Exhaust the window with wrong passwords until throttled.
    let throttled = false;
    for (let i = 0; i < 10 && !throttled; i++) {
      throttled = (await tryLogin('wrong-' + i)).status === 429;
    }
    expect(throttled, 'the IP should be throttled after a burst').toBe(true);

    // The right password, while throttled, must STILL be refused — the limiter is checked before
    // the bcrypt verify, so a valid credential does not buy a way around a brute-force lockout.
    expect(
      (await tryLogin(ADMIN_PASSWORD)).status,
      'a throttled IP is refused even with the correct password'
    ).toBe(429);
  });

  test('the throttle is gated: with the limiter disabled, a burst is NOT rate-limited', async ({
    api,
    adminToken,
  }) => {
    // The mirror of the above — proves the toggle actually gates the behaviour (and documents that
    // the e2e default really is "off", which is why every OTHER admin test can hammer login freely).
    await api.patchConfig(adminToken, {
      rate_limits_enabled: 'true',
      admin_login_rate_enabled: 'false',
    });

    const statuses: number[] = [];
    for (let i = 0; i < 8; i++) statuses.push((await tryLogin('wrong-' + i)).status);

    expect(
      statuses.every((s) => s === 401),
      'with admin_login_rate_enabled=false no attempt should be 429'
    ).toBe(true);
  });
});