fix(federation): harness round-2 — email validation + host-side URL rewrite

- Bug-1: replace whitespace in admin email local-part (was breaking @IsEmail) - Bug-2: rewrite enrollment URL to use host-accessible base in seed.ts (in-cluster URL not resolvable from host) - Bug-3: correct README Known Limitations section - eslint.config.mjs: add tools/federation-harness/*.ts to allowDefaultProject so pre-commit hook can lint harness scripts Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
fix(federation): harness CRIT bugs — admin bootstrap auth + peer FK + boot deadline (review remediation)
2026-04-23 21:54:46 -05:00 · 2026-04-23 21:54:46 -05:00 · 2026-04-23 21:54:46 -05:00
23 changed files with 142 additions and 2468 deletions
--- a/apps/gateway/package.json
+++ b/apps/gateway/package.json
@@ -73,7 +73,6 @@
    "rxjs": "^7.8.0",
    "socket.io": "^4.8.0",
    "uuid": "^11.0.0",
    "undici": "^7.24.6",
    "zod": "^4.3.6"
  },
  "devDependencies": {
--- a/apps/gateway/src/database/database.module.ts
+++ b/apps/gateway/src/database/database.module.ts
@@ -1,21 +1,8 @@
 import { mkdirSync } from 'node:fs';
 import { homedir } from 'node:os';
 import { join } from 'node:path';
-import {
+import { Global, Inject, Module, type OnApplicationShutdown } from '@nestjs/common';
-  Global,
+import { createDb, createPgliteDb, type Db, type DbHandle } from '@mosaicstack/db';
  Inject,
  Logger,
  Module,
  type OnApplicationShutdown,
  type OnModuleInit,
 } from '@nestjs/common';
 import {
  createDb,
  createPgliteDb,
  runPgliteMigrations,
  type Db,
  type DbHandle,
 } from '@mosaicstack/db';
 import { createStorageAdapter, type StorageAdapter } from '@mosaicstack/storage';
 import type { MosaicConfig } from '@mosaicstack/config';
 import { MOSAIC_CONFIG } from '../config/config.module.js';
@@ -52,37 +39,12 @@ export const STORAGE_ADAPTER = 'STORAGE_ADAPTER';
  ],
  exports: [DB, STORAGE_ADAPTER],
 })
-export class DatabaseModule implements OnApplicationShutdown, OnModuleInit {
+export class DatabaseModule implements OnApplicationShutdown {
  private readonly logger = new Logger(DatabaseModule.name);
  constructor(
    @Inject(DB_HANDLE) private readonly handle: DbHandle,
    @Inject(STORAGE_ADAPTER) private readonly storageAdapter: StorageAdapter,
    @Inject(MOSAIC_CONFIG) private readonly config: MosaicConfig,
  ) {}
  // Migrations must complete before any module that injects DB starts serving
  // requests. NestJS awaits onModuleInit before app.listen(), and modules that
  // inject DB are initialized after this one — so all DB-dependent code sees a
  // populated schema before the first HTTP request lands.
  //
  // Local (PGlite) tier: we run gateway-DB migrations explicitly here. The
  // storage adapter writes to a separate PGlite directory and only manages its
  // own KV tables, so we still call its migrate() afterwards.
  //
  // Postgres tier: PostgresAdapter.migrate() already calls runMigrations() on
  // the same DATABASE_URL, so a single call covers both the gateway DB and
  // the storage tables. We deliberately do NOT call runMigrations() here to
  // avoid opening a second short-lived connection and doubling startup cost.
  async onModuleInit(): Promise<void> {
    if (this.config.tier === 'local') {
      this.logger.log('Applying PGlite schema migrations...');
      await runPgliteMigrations(this.handle);
    }
    this.logger.log(`Initializing storage adapter (${this.storageAdapter.name})...`);
    await this.storageAdapter.migrate();
  }
  async onApplicationShutdown(): Promise<void> {
    await Promise.all([this.handle.close(), this.storageAdapter.close()]);
  }
--- a/apps/gateway/src/federation/client/tests/federation-client.service.spec.ts
+++ b/apps/gateway/src/federation/client/tests/federation-client.service.spec.ts
@@ -1,553 +0,0 @@
 /**
 * Unit tests for FederationClientService (FED-M3-08).
 *
 * HTTP mocking strategy:
 *   undici MockAgent is used to intercept outbound HTTP requests.  The service
 *   uses `undici.fetch` with a `dispatcher` option, so MockAgent is set as the
 *   global dispatcher and all requests flow through it.
 *
 *   Because the service builds one `undici.Agent` per peer and passes it as
 *   the dispatcher on every fetch call, we cannot intercept at the Agent level
 *   in unit tests without significant refactoring.  Instead, we set the global
 *   dispatcher to a MockAgent and override the service's `doRequest` indirection
 *   by spying on the internal fetch call.
 *
 *   For the cert/key wiring, we use the real `sealClientKey` function from
 *   peer-key.util.ts with a test secret — no stubs.
 *
 * Sealed-key setup:
 *   Each test (or beforeAll) calls `sealClientKey(TEST_PRIVATE_KEY_PEM)` with
 *   BETTER_AUTH_SECRET set to a deterministic test value so that
 *   `unsealClientKey` in the service recovers the original PEM.
 */
 import 'reflect-metadata';
 import { describe, it, expect, vi, beforeEach, afterEach, beforeAll, afterAll } from 'vitest';
 import { MockAgent, setGlobalDispatcher, getGlobalDispatcher } from 'undici';
 import type { Dispatcher } from 'undici';
 import { writeFileSync, unlinkSync } from 'node:fs';
 import { tmpdir } from 'node:os';
 import { join } from 'node:path';
 import type { Db } from '@mosaicstack/db';
 import { FederationClientService, FederationClientError } from '../federation-client.service.js';
 import { sealClientKey } from '../../peer-key.util.js';
 // ---------------------------------------------------------------------------
 // Test constants
 // ---------------------------------------------------------------------------
 const TEST_SECRET = 'test-secret-for-federation-client-spec-only';
 const PEER_ID = 'aaaaaaaa-aaaa-aaaa-aaaa-aaaaaaaaaaaa';
 const ENDPOINT = 'https://peer.example.com';
 // Minimal valid RSA/EC private key PEM — does NOT need to be a real key for
 // unit tests because we only verify it round-trips through seal/unseal, not
 // that it actually negotiates TLS (MockAgent handles that).
 const TEST_PRIVATE_KEY_PEM = `-----BEGIN PRIVATE KEY-----
 MIIEvQIBADANBgkqhkiG9w0BAQEFAASCBKcwggSjAgEAAoIBAQDummyKeyForTests
 -----END PRIVATE KEY-----`;
 // Minimal self-signed cert PEM (dummy — only used for mTLS Agent construction)
 const TEST_CERT_PEM = `-----BEGIN CERTIFICATE-----
 MIIBdummyCertForFederationClientTests==
 -----END CERTIFICATE-----`;
 const TEST_CERT_SERIAL = 'ABCDEF1234567890';
 // ---------------------------------------------------------------------------
 // Sealed key (computed once in beforeAll)
 // ---------------------------------------------------------------------------
 let SEALED_KEY: string;
 // Path to a stub Step-CA root cert file written in beforeAll. The cert is never
 // actually used to negotiate TLS in unit tests (MockAgent + spy on resolveEntry
 // short-circuit the network), but loadStepCaRoot() requires the file to exist.
 const STUB_CA_PEM_PATH = join(tmpdir(), 'federation-client-spec-ca.pem');
 const STUB_CA_PEM = `-----BEGIN CERTIFICATE-----
 MIIBdummyCAforFederationClientSpecOnly==
 -----END CERTIFICATE-----
 `;
 // ---------------------------------------------------------------------------
 // Peer row factory
 // ---------------------------------------------------------------------------
 function makePeerRow(overrides: Partial<Record<string, unknown>> = {}) {
  return {
    id: PEER_ID,
    commonName: 'peer-example-com',
    displayName: 'Test Peer',
    certPem: TEST_CERT_PEM,
    certSerial: TEST_CERT_SERIAL,
    certNotAfter: new Date('2030-01-01T00:00:00Z'),
    clientKeyPem: SEALED_KEY,
    state: 'active' as const,
    endpointUrl: ENDPOINT,
    lastSeenAt: null,
    createdAt: new Date('2026-01-01T00:00:00Z'),
    revokedAt: null,
    ...overrides,
  };
 }
 // ---------------------------------------------------------------------------
 // Mock DB builder
 // ---------------------------------------------------------------------------
 function makeDb(selectRows: unknown[] = [makePeerRow()]): Db {
  const limitSelect = vi.fn().mockResolvedValue(selectRows);
  const whereSelect = vi.fn().mockReturnValue({ limit: limitSelect });
  const fromSelect = vi.fn().mockReturnValue({ where: whereSelect });
  const selectMock = vi.fn().mockReturnValue({ from: fromSelect });
  return {
    select: selectMock,
    insert: vi.fn(),
    update: vi.fn(),
    delete: vi.fn(),
    transaction: vi.fn(),
  } as unknown as Db;
 }
 // ---------------------------------------------------------------------------
 // Helpers for MockAgent HTTP interception
 // ---------------------------------------------------------------------------
 /**
 * Create a MockAgent + MockPool for the peer endpoint, set it as the global
 * dispatcher, and return both for per-test configuration.
 */
 function makeMockAgent() {
  const mockAgent = new MockAgent({ connections: 1 });
  mockAgent.disableNetConnect();
  setGlobalDispatcher(mockAgent);
  const pool = mockAgent.get(ENDPOINT);
  return { mockAgent, pool };
 }
 /**
 * Build a FederationClientService with a mock DB and a spy on the internal
 * fetch so we can intercept at the HTTP layer via MockAgent.
 *
 * The service calls `fetch(url, { dispatcher: agent })` where `agent` is the
 * mTLS undici.Agent built from the peer's cert+key.  To make MockAgent work,
 * we need the fetch dispatcher to be the MockAgent, not the per-peer Agent.
 *
 * Strategy: we replace the private `resolveEntry` result's `agent` field with
 * the MockAgent's pool, so fetch uses our interceptor.  We do this by spying
 * on `resolveEntry` and returning a controlled entry.
 */
 function makeService(db: Db, mockPool: Dispatcher): FederationClientService {
  const svc = new FederationClientService(db);
  // Override resolveEntry to inject MockAgent pool as the dispatcher
  vi.spyOn(
    svc as unknown as { resolveEntry: (peerId: string) => Promise<unknown> },
    'resolveEntry',
  ).mockImplementation(async (_peerId: string) => {
    // Still call DB (via the real logic) to exercise peer validation,
    // but return mock pool as the agent.
    // For simplicity in unit tests, directly return a controlled entry.
    return {
      agent: mockPool,
      endpointUrl: ENDPOINT,
      certPem: TEST_CERT_PEM,
      certSerial: TEST_CERT_SERIAL,
    };
  });
  return svc;
 }
 // ---------------------------------------------------------------------------
 // Test setup
 // ---------------------------------------------------------------------------
 let originalDispatcher: Dispatcher;
 beforeAll(() => {
  // Seal the test key once — requires BETTER_AUTH_SECRET
  const saved = process.env['BETTER_AUTH_SECRET'];
  process.env['BETTER_AUTH_SECRET'] = TEST_SECRET;
  try {
    SEALED_KEY = sealClientKey(TEST_PRIVATE_KEY_PEM);
  } finally {
    if (saved === undefined) {
      delete process.env['BETTER_AUTH_SECRET'];
    } else {
      process.env['BETTER_AUTH_SECRET'] = saved;
    }
  }
  writeFileSync(STUB_CA_PEM_PATH, STUB_CA_PEM, 'utf8');
 });
 afterAll(() => {
  try {
    unlinkSync(STUB_CA_PEM_PATH);
  } catch {
    // best-effort cleanup
  }
 });
 beforeEach(() => {
  originalDispatcher = getGlobalDispatcher();
  process.env['BETTER_AUTH_SECRET'] = TEST_SECRET;
  process.env['STEP_CA_ROOT_CERT_PATH'] = STUB_CA_PEM_PATH;
 });
 afterEach(() => {
  setGlobalDispatcher(originalDispatcher);
  vi.restoreAllMocks();
  delete process.env['BETTER_AUTH_SECRET'];
  delete process.env['STEP_CA_ROOT_CERT_PATH'];
 });
 // ---------------------------------------------------------------------------
 // Helpers
 // ---------------------------------------------------------------------------
 /** Successful list response body */
 const LIST_BODY = {
  items: [{ id: '1', title: 'Task One' }],
  nextCursor: undefined,
  _partial: false,
 };
 /** Successful get response body */
 const GET_BODY = {
  item: { id: '1', title: 'Task One' },
  _partial: false,
 };
 /** Successful capabilities response body */
 const CAP_BODY = {
  resources: ['tasks'],
  excluded_resources: [],
  max_rows_per_query: 100,
  supported_verbs: ['list', 'get', 'capabilities'] as const,
 };
 // ---------------------------------------------------------------------------
 // Tests
 // ---------------------------------------------------------------------------
 describe('FederationClientService', () => {
  // ─── Successful verb calls ─────────────────────────────────────────────────
  describe('list()', () => {
    it('returns parsed typed response on success', async () => {
      const db = makeDb();
      const { mockAgent, pool } = makeMockAgent();
      const svc = makeService(db, pool);
      pool
        .intercept({
          path: '/api/federation/v1/list/tasks',
          method: 'POST',
        })
        .reply(200, LIST_BODY, { headers: { 'content-type': 'application/json' } });
      const result = await svc.list(PEER_ID, 'tasks', {});
      expect(result.items).toHaveLength(1);
      expect(result.items[0]).toMatchObject({ id: '1', title: 'Task One' });
      await mockAgent.close();
    });
  });
  describe('get()', () => {
    it('returns parsed typed response on success', async () => {
      const db = makeDb();
      const { mockAgent, pool } = makeMockAgent();
      const svc = makeService(db, pool);
      pool
        .intercept({
          path: '/api/federation/v1/get/tasks/1',
          method: 'POST',
        })
        .reply(200, GET_BODY, { headers: { 'content-type': 'application/json' } });
      const result = await svc.get(PEER_ID, 'tasks', '1', {});
      expect(result.item).toMatchObject({ id: '1', title: 'Task One' });
      await mockAgent.close();
    });
  });
  describe('capabilities()', () => {
    it('returns parsed capabilities response on success', async () => {
      const db = makeDb();
      const { mockAgent, pool } = makeMockAgent();
      const svc = makeService(db, pool);
      pool
        .intercept({
          path: '/api/federation/v1/capabilities',
          method: 'GET',
        })
        .reply(200, CAP_BODY, { headers: { 'content-type': 'application/json' } });
      const result = await svc.capabilities(PEER_ID);
      expect(result.resources).toContain('tasks');
      expect(result.max_rows_per_query).toBe(100);
      await mockAgent.close();
    });
  });
  // ─── HTTP error surfaces ──────────────────────────────────────────────────
  describe('non-2xx responses', () => {
    it('surfaces 403 as FederationClientError({ status: 403, code: "FORBIDDEN" })', async () => {
      const db = makeDb();
      const { mockAgent, pool } = makeMockAgent();
      const svc = makeService(db, pool);
      pool.intercept({ path: '/api/federation/v1/list/tasks', method: 'POST' }).reply(
        403,
        { error: { code: 'forbidden', message: 'Access denied' } },
        {
          headers: { 'content-type': 'application/json' },
        },
      );
      await expect(svc.list(PEER_ID, 'tasks', {})).rejects.toMatchObject({
        status: 403,
        code: 'FORBIDDEN',
        peerId: PEER_ID,
      });
      await mockAgent.close();
    });
    it('surfaces 404 as FederationClientError({ status: 404, code: "HTTP_404" })', async () => {
      const db = makeDb();
      const { mockAgent, pool } = makeMockAgent();
      const svc = makeService(db, pool);
      pool.intercept({ path: '/api/federation/v1/get/tasks/999', method: 'POST' }).reply(
        404,
        { error: { code: 'not_found', message: 'Not found' } },
        {
          headers: { 'content-type': 'application/json' },
        },
      );
      await expect(svc.get(PEER_ID, 'tasks', '999', {})).rejects.toMatchObject({
        status: 404,
        code: 'HTTP_404',
        peerId: PEER_ID,
      });
      await mockAgent.close();
    });
  });
  // ─── Network error ─────────────────────────────────────────────────────────
  describe('network errors', () => {
    it('surfaces network error as FederationClientError({ code: "NETWORK" })', async () => {
      const db = makeDb();
      const { mockAgent, pool } = makeMockAgent();
      const svc = makeService(db, pool);
      pool
        .intercept({ path: '/api/federation/v1/capabilities', method: 'GET' })
        .replyWithError(new Error('ECONNREFUSED'));
      await expect(svc.capabilities(PEER_ID)).rejects.toMatchObject({
        code: 'NETWORK',
        peerId: PEER_ID,
      });
      await mockAgent.close();
    });
  });
  // ─── Invalid response body ─────────────────────────────────────────────────
  describe('invalid response body', () => {
    it('surfaces as FederationClientError({ code: "INVALID_RESPONSE" }) when body shape is wrong', async () => {
      const db = makeDb();
      const { mockAgent, pool } = makeMockAgent();
      const svc = makeService(db, pool);
      // capabilities returns wrong shape (missing required fields)
      pool
        .intercept({ path: '/api/federation/v1/capabilities', method: 'GET' })
        .reply(200, { totally: 'wrong' }, { headers: { 'content-type': 'application/json' } });
      await expect(svc.capabilities(PEER_ID)).rejects.toMatchObject({
        code: 'INVALID_RESPONSE',
        peerId: PEER_ID,
      });
      await mockAgent.close();
    });
  });
  // ─── Peer DB validation ────────────────────────────────────────────────────
  describe('peer validation (without resolveEntry spy)', () => {
    /**
     * These tests exercise the real `resolveEntry` path — no spy on resolveEntry.
     */
    it('throws PEER_NOT_FOUND when peer is not in DB', async () => {
      // DB returns empty array (peer not found)
      const db = makeDb([]);
      const svc = new FederationClientService(db);
      await expect(svc.capabilities(PEER_ID)).rejects.toMatchObject({
        code: 'PEER_NOT_FOUND',
        peerId: PEER_ID,
      });
    });
    it('throws PEER_INACTIVE when peer state is not "active"', async () => {
      const db = makeDb([makePeerRow({ state: 'suspended' })]);
      const svc = new FederationClientService(db);
      await expect(svc.capabilities(PEER_ID)).rejects.toMatchObject({
        code: 'PEER_INACTIVE',
        peerId: PEER_ID,
      });
    });
  });
  // ─── Cache behaviour ───────────────────────────────────────────────────────
  describe('cache behaviour', () => {
    it('hits cache on second call — only one DB lookup happens', async () => {
      // Verify cache by calling the private resolveEntry directly twice and
      // asserting the DB was queried only once. This avoids the HTTP layer,
      // which would require either a real network or per-peer Agent rewiring
      // that the cache invariant doesn't depend on.
      const db = makeDb();
      const selectSpy = vi.spyOn(db, 'select');
      const svc = new FederationClientService(db);
      const resolveEntry = (
        svc as unknown as { resolveEntry: (peerId: string) => Promise<unknown> }
      ).resolveEntry.bind(svc);
      const first = await resolveEntry(PEER_ID);
      const second = await resolveEntry(PEER_ID);
      expect(first).toBe(second);
      expect(selectSpy).toHaveBeenCalledTimes(1);
    });
    it('serializes concurrent resolveEntry calls — only one DB lookup', async () => {
      const db = makeDb();
      const selectSpy = vi.spyOn(db, 'select');
      const svc = new FederationClientService(db);
      const resolveEntry = (
        svc as unknown as {
          resolveEntry: (peerId: string) => Promise<unknown>;
        }
      ).resolveEntry.bind(svc);
      const [a, b] = await Promise.all([resolveEntry(PEER_ID), resolveEntry(PEER_ID)]);
      expect(a).toBe(b);
      expect(selectSpy).toHaveBeenCalledTimes(1);
    });
    it('flushPeer destroys the evicted Agent so old TLS connections close', async () => {
      const db = makeDb();
      const svc = new FederationClientService(db);
      const resolveEntry = (
        svc as unknown as {
          resolveEntry: (peerId: string) => Promise<{ agent: { destroy: () => Promise<void> } }>;
        }
      ).resolveEntry.bind(svc);
      const entry = await resolveEntry(PEER_ID);
      const destroySpy = vi.spyOn(entry.agent, 'destroy').mockResolvedValue();
      svc.flushPeer(PEER_ID);
      expect(destroySpy).toHaveBeenCalledTimes(1);
    });
    it('flushPeer() invalidates cache — next call re-reads DB', async () => {
      const db = makeDb();
      const { mockAgent, pool } = makeMockAgent();
      const svc = makeService(db, pool);
      pool
        .intercept({ path: '/api/federation/v1/capabilities', method: 'GET' })
        .reply(200, CAP_BODY, { headers: { 'content-type': 'application/json' } })
        .times(2);
      // First call — populates cache (via mock resolveEntry)
      await svc.capabilities(PEER_ID);
      // Flush the cache
      svc.flushPeer(PEER_ID);
      // The spy on resolveEntry is still active — check it's called again after flush
      const resolveEntrySpy = vi.spyOn(
        svc as unknown as { resolveEntry: (peerId: string) => Promise<unknown> },
        'resolveEntry',
      );
      // Second call after flush — should call resolveEntry again
      await svc.capabilities(PEER_ID);
      // resolveEntry should have been called once after we started spying (post-flush)
      expect(resolveEntrySpy).toHaveBeenCalledTimes(1);
      await mockAgent.close();
    });
  });
  // ─── loadStepCaRoot env-var guard ─────────────────────────────────────────
  describe('loadStepCaRoot() env-var guard', () => {
    it('throws PEER_MISCONFIGURED when STEP_CA_ROOT_CERT_PATH is not set', async () => {
      delete process.env['STEP_CA_ROOT_CERT_PATH'];
      const db = makeDb();
      const svc = new FederationClientService(db);
      const resolveEntry = (
        svc as unknown as {
          resolveEntry: (peerId: string) => Promise<unknown>;
        }
      ).resolveEntry.bind(svc);
      await expect(resolveEntry(PEER_ID)).rejects.toMatchObject({
        code: 'PEER_MISCONFIGURED',
      });
    });
  });
  // ─── FederationClientError class ──────────────────────────────────────────
  describe('FederationClientError', () => {
    it('is instanceof Error and FederationClientError', () => {
      const err = new FederationClientError({
        code: 'PEER_NOT_FOUND',
        message: 'test',
        peerId: PEER_ID,
      });
      expect(err).toBeInstanceOf(Error);
      expect(err).toBeInstanceOf(FederationClientError);
      expect(err.name).toBe('FederationClientError');
    });
    it('carries status, code, and peerId', () => {
      const err = new FederationClientError({
        status: 403,
        code: 'FORBIDDEN',
        message: 'forbidden',
        peerId: PEER_ID,
      });
      expect(err.status).toBe(403);
      expect(err.code).toBe('FORBIDDEN');
      expect(err.peerId).toBe(PEER_ID);
    });
  });
 });
--- a/apps/gateway/src/federation/client/federation-client.service.ts
+++ b/apps/gateway/src/federation/client/federation-client.service.ts
@@ -1,500 +0,0 @@
 /**
 * FederationClientService — outbound mTLS client for federation requests (FED-M3-08).
 *
 * Dials peer gateways over mTLS using the cert+sealed-key stored in `federation_peers`,
 * invokes federation verbs (list / get / capabilities), and surfaces all failure modes
 * as typed `FederationClientError` instances.
 *
 * ## Error code taxonomy
 *
 * | Code               | When                                                          |
 * | ------------------ | ------------------------------------------------------------- |
 * | PEER_NOT_FOUND     | No row in federation_peers for the given peerId               |
 * | PEER_INACTIVE      | Peer row exists but state !== 'active'                        |
 * | PEER_MISCONFIGURED | Peer row is active but missing endpointUrl or clientKeyPem    |
 * | NETWORK            | undici threw a connection / TLS / timeout error               |
 * | HTTP_{status}      | Peer returned a non-2xx response (e.g. HTTP_403, HTTP_404)    |
 * | FORBIDDEN          | Peer returned 403 (convenience alias alongside HTTP_403)      |
 * | INVALID_RESPONSE   | Response body failed Zod schema validation                    |
 *
 * ## Cache strategy
 *
 * Per-peer `undici.Agent` instances are cached in a `Map<peerId, AgentCacheEntry>` for
 * the lifetime of the service instance.  The cache is keyed on peerId (UUID).
 *
 * Cache invalidation:
 *  - `flushPeer(peerId)` — removes the entry immediately.  M5/M6 MUST call this on
 *    cert rotation or peer revocation events so the next request re-reads the DB and
 *    builds a fresh TLS Agent with the new cert material.
 *  - On cache miss: re-reads the DB, checks state === 'active', rebuilds Agent.
 *
 * Cache does NOT auto-expire.  The service is expected to be a singleton scoped to the
 * NestJS application lifecycle; flushing on revocation/rotation is the only invalidation
 * path by design (avoids redundant DB round-trips on the hot path).
 */
 import { Injectable, Inject, Logger } from '@nestjs/common';
 import { readFileSync } from 'node:fs';
 import { Agent, fetch as undiciFetch } from 'undici';
 import type { Dispatcher } from 'undici';
 import { z } from 'zod';
 import { type Db, eq, federationPeers } from '@mosaicstack/db';
 import {
  FederationListResponseSchema,
  FederationGetResponseSchema,
  FederationCapabilitiesResponseSchema,
  FederationErrorEnvelopeSchema,
  type FederationListResponse,
  type FederationGetResponse,
  type FederationCapabilitiesResponse,
 } from '@mosaicstack/types';
 import { DB } from '../../database/database.module.js';
 import { unsealClientKey } from '../peer-key.util.js';
 // ---------------------------------------------------------------------------
 // Error taxonomy
 // ---------------------------------------------------------------------------
 /**
 * Client-side error code set.  Distinct from the server-side `FederationErrorCode`
 * (which lives in `@mosaicstack/types`) because the client has additional failure
 * modes (PEER_NOT_FOUND, PEER_INACTIVE, PEER_MISCONFIGURED, NETWORK) that the
 * server never emits.
 */
 export type FederationClientErrorCode =
  | 'PEER_NOT_FOUND'
  | 'PEER_INACTIVE'
  | 'PEER_MISCONFIGURED'
  | 'NETWORK'
  | 'FORBIDDEN'
  | 'INVALID_RESPONSE'
  | `HTTP_${number}`;
 export interface FederationClientErrorOptions {
  status?: number;
  code: FederationClientErrorCode;
  message: string;
  peerId: string;
  cause?: unknown;
 }
 /**
 * Thrown by FederationClientService on every failure path.
 * Callers can dispatch on `error.code` for programmatic handling.
 */
 export class FederationClientError extends Error {
  readonly status?: number;
  readonly code: FederationClientErrorCode;
  readonly peerId: string;
  readonly cause?: unknown;
  constructor(opts: FederationClientErrorOptions) {
    super(opts.message);
    this.name = 'FederationClientError';
    this.status = opts.status;
    this.code = opts.code;
    this.peerId = opts.peerId;
    this.cause = opts.cause;
  }
 }
 // ---------------------------------------------------------------------------
 // Internal cache types
 // ---------------------------------------------------------------------------
 interface AgentCacheEntry {
  agent: Agent;
  endpointUrl: string;
  certPem: string;
  certSerial: string;
 }
 // ---------------------------------------------------------------------------
 // Service
 // ---------------------------------------------------------------------------
@Injectable()
 export class FederationClientService {
  private readonly logger = new Logger(FederationClientService.name);
  /**
   * Per-peer undici Agent cache.
   * Key = peerId (UUID string).
   *
   * Values are either a resolved `AgentCacheEntry` or an in-flight
   * `Promise<AgentCacheEntry>` (promise-cache pattern).  Storing the promise
   * prevents duplicate DB lookups and duplicate key-unseal operations when two
   * requests for the same peer arrive before the first build completes.
   *
   * Flush via `flushPeer(peerId)` on cert rotation / peer revocation (M5/M6).
   */
  private readonly cache = new Map<string, AgentCacheEntry | Promise<AgentCacheEntry>>();
  /**
   * Step-CA root cert PEM, loaded once from `STEP_CA_ROOT_CERT_PATH`.
   * Used as the trust anchor for peer server certificates so federation TLS is
   * pinned to our PKI, not the public trust store. Lazily loaded on first use
   * so unit tests that don't exercise the agent path can run without the env var.
   */
  private cachedCaPem: string | null = null;
  constructor(@Inject(DB) private readonly db: Db) {}
  // -------------------------------------------------------------------------
  // Public verb API
  // -------------------------------------------------------------------------
  /**
   * Invoke the `list` verb on a remote peer.
   *
   * @param peerId   UUID of the peer row in `federation_peers`.
   * @param resource Resource path, e.g. "tasks".
   * @param request  Free-form body sent as JSON in the POST body.
   * @returns Parsed `FederationListResponse<T>`.
   */
  async list<T>(
    peerId: string,
    resource: string,
    request: Record<string, unknown>,
  ): Promise<FederationListResponse<T>> {
    const { endpointUrl, agent } = await this.resolveEntry(peerId);
    const url = `${endpointUrl}/api/federation/v1/list/${encodeURIComponent(resource)}`;
    const body = await this.doPost(peerId, url, agent, request);
    return this.parseWith<FederationListResponse<T>>(
      peerId,
      body,
      FederationListResponseSchema(z.unknown()),
    );
  }
  /**
   * Invoke the `get` verb on a remote peer.
   *
   * @param peerId   UUID of the peer row in `federation_peers`.
   * @param resource Resource path, e.g. "tasks".
   * @param id       Resource identifier.
   * @param request  Free-form body sent as JSON in the POST body.
   * @returns Parsed `FederationGetResponse<T>`.
   */
  async get<T>(
    peerId: string,
    resource: string,
    id: string,
    request: Record<string, unknown>,
  ): Promise<FederationGetResponse<T>> {
    const { endpointUrl, agent } = await this.resolveEntry(peerId);
    const url = `${endpointUrl}/api/federation/v1/get/${encodeURIComponent(resource)}/${encodeURIComponent(id)}`;
    const body = await this.doPost(peerId, url, agent, request);
    return this.parseWith<FederationGetResponse<T>>(
      peerId,
      body,
      FederationGetResponseSchema(z.unknown()),
    );
  }
  /**
   * Invoke the `capabilities` verb on a remote peer.
   *
   * @param peerId UUID of the peer row in `federation_peers`.
   * @returns Parsed `FederationCapabilitiesResponse`.
   */
  async capabilities(peerId: string): Promise<FederationCapabilitiesResponse> {
    const { endpointUrl, agent } = await this.resolveEntry(peerId);
    const url = `${endpointUrl}/api/federation/v1/capabilities`;
    const body = await this.doGet(peerId, url, agent);
    return this.parseWith<FederationCapabilitiesResponse>(
      peerId,
      body,
      FederationCapabilitiesResponseSchema,
    );
  }
  // -------------------------------------------------------------------------
  // Cache management
  // -------------------------------------------------------------------------
  /**
   * Flush the cached Agent for a specific peer.
   *
   * M5/M6 MUST call this on:
   *  - cert rotation events (so new cert material is picked up)
   *  - peer revocation events (so future requests fail at PEER_INACTIVE)
   *
   * After flushing, the next call to `list`, `get`, or `capabilities` for
   * this peer will re-read the DB and rebuild the Agent.
   */
  flushPeer(peerId: string): void {
    const entry = this.cache.get(peerId);
    if (entry === undefined) {
      return;
    }
    this.cache.delete(peerId);
    if (!(entry instanceof Promise)) {
      // best-effort destroy; promise-cached entries skip destroy because
      // the in-flight build owns its own Agent which will be GC'd when the
      // owning request handles the rejection from the cache miss
      entry.agent.destroy().catch(() => {
        // intentionally ignored — destroy errors are not actionable
      });
    }
    this.logger.log(`Cache flushed for peer ${peerId}`);
  }
  // -------------------------------------------------------------------------
  // Internal helpers
  // -------------------------------------------------------------------------
  /**
   * Load and cache the Step-CA root cert PEM from `STEP_CA_ROOT_CERT_PATH`.
   * Throws `FederationClientError` if the env var is unset or the file cannot
   * be read — mTLS to a peer without a pinned trust anchor would silently
   * fall back to the public trust store.
   */
  private loadStepCaRoot(): string {
    if (this.cachedCaPem !== null) {
      return this.cachedCaPem;
    }
    const path = process.env['STEP_CA_ROOT_CERT_PATH'];
    if (!path) {
      throw new FederationClientError({
        code: 'PEER_MISCONFIGURED',
        message: 'STEP_CA_ROOT_CERT_PATH is not set; refusing to dial peer without pinned CA trust',
        peerId: '',
      });
    }
    try {
      const pem = readFileSync(path, 'utf8');
      this.cachedCaPem = pem;
      return pem;
    } catch (err) {
      throw new FederationClientError({
        code: 'PEER_MISCONFIGURED',
        message: `Failed to read STEP_CA_ROOT_CERT_PATH (${path})`,
        peerId: '',
        cause: err,
      });
    }
  }
  /**
   * Resolve the cache entry for a peer, reading DB on miss.
   *
   * Uses a promise-cache pattern: concurrent callers for the same uncached
   * `peerId` all `await` the same in-flight `Promise<AgentCacheEntry>` so
   * only one DB lookup and one key-unseal ever runs per peer per cache miss.
   * The promise is replaced with the concrete entry on success, or deleted on
   * rejection so a transient error does not poison the cache permanently.
   *
   * Throws `FederationClientError` with appropriate code if the peer is not
   * found, is inactive, or is missing required fields.
   */
  private async resolveEntry(peerId: string): Promise<AgentCacheEntry> {
    const cached = this.cache.get(peerId);
    if (cached) {
      return cached; // Promise or concrete entry — both are awaitable
    }
    const inflight = this.buildEntry(peerId).then(
      (entry) => {
        this.cache.set(peerId, entry); // replace promise with concrete value
        return entry;
      },
      (err: unknown) => {
        this.cache.delete(peerId); // don't poison the cache with a rejected promise
        throw err;
      },
    );
    this.cache.set(peerId, inflight);
    return inflight;
  }
  /**
   * Build the `AgentCacheEntry` for a peer by reading the DB, validating the
   * peer's state, unsealing the private key, and constructing the mTLS Agent.
   *
   * Throws `FederationClientError` with appropriate code if the peer is not
   * found, is inactive, or is missing required fields.
   */
  private async buildEntry(peerId: string): Promise<AgentCacheEntry> {
    // DB lookup
    const [peer] = await this.db
      .select()
      .from(federationPeers)
      .where(eq(federationPeers.id, peerId))
      .limit(1);
    if (!peer) {
      throw new FederationClientError({
        code: 'PEER_NOT_FOUND',
        message: `Federation peer ${peerId} not found`,
        peerId,
      });
    }
    if (peer.state !== 'active') {
      throw new FederationClientError({
        code: 'PEER_INACTIVE',
        message: `Federation peer ${peerId} is not active (state: ${peer.state})`,
        peerId,
      });
    }
    if (!peer.endpointUrl || !peer.clientKeyPem) {
      throw new FederationClientError({
        code: 'PEER_MISCONFIGURED',
        message: `Federation peer ${peerId} is missing endpointUrl or clientKeyPem`,
        peerId,
      });
    }
    // Unseal the private key
    let privateKeyPem: string;
    try {
      privateKeyPem = unsealClientKey(peer.clientKeyPem);
    } catch (err) {
      throw new FederationClientError({
        code: 'PEER_MISCONFIGURED',
        message: `Failed to unseal client key for peer ${peerId}`,
        peerId,
        cause: err,
      });
    }
    // Build mTLS agent — pin trust to Step-CA root so we never accept
    // a peer cert signed by a public CA (defense against MITM with a
    // publicly-trusted DV cert for the peer's hostname).
    const agent = new Agent({
      connect: {
        cert: peer.certPem,
        key: privateKeyPem,
        ca: this.loadStepCaRoot(),
        // rejectUnauthorized: true is the undici default for HTTPS
      },
    });
    const entry: AgentCacheEntry = {
      agent,
      endpointUrl: peer.endpointUrl,
      certPem: peer.certPem,
      certSerial: peer.certSerial,
    };
    this.logger.log(`Agent cached for peer ${peerId} (serial: ${peer.certSerial})`);
    return entry;
  }
  /**
   * Execute a POST request with a JSON body.
   * Returns the parsed response body as an unknown value.
   * Throws `FederationClientError` on network errors and non-2xx responses.
   */
  private async doPost(
    peerId: string,
    url: string,
    agent: Dispatcher,
    body: Record<string, unknown>,
  ): Promise<unknown> {
    return this.doRequest(peerId, url, agent, {
      method: 'POST',
      headers: { 'Content-Type': 'application/json' },
      body: JSON.stringify(body),
    });
  }
  /**
   * Execute a GET request.
   * Returns the parsed response body as an unknown value.
   * Throws `FederationClientError` on network errors and non-2xx responses.
   */
  private async doGet(peerId: string, url: string, agent: Dispatcher): Promise<unknown> {
    return this.doRequest(peerId, url, agent, { method: 'GET' });
  }
  private async doRequest(
    peerId: string,
    url: string,
    agent: Dispatcher,
    init: { method: string; headers?: Record<string, string>; body?: string },
  ): Promise<unknown> {
    let response: Awaited<ReturnType<typeof undiciFetch>>;
    try {
      response = await undiciFetch(url, {
        ...init,
        dispatcher: agent,
      });
    } catch (err) {
      throw new FederationClientError({
        code: 'NETWORK',
        message: `Network error calling peer ${peerId} at ${url}: ${err instanceof Error ? err.message : String(err)}`,
        peerId,
        cause: err,
      });
    }
    const rawBody = await response.text().catch(() => '');
    if (!response.ok) {
      const status = response.status;
      // Attempt to parse as federation error envelope
      let serverMessage = `HTTP ${status}`;
      try {
        const json: unknown = JSON.parse(rawBody);
        const result = FederationErrorEnvelopeSchema.safeParse(json);
        if (result.success) {
          serverMessage = result.data.error.message;
        }
      } catch {
        // Not valid JSON or not a federation envelope — use generic message
      }
      // Specific code for 403 (most actionable for callers); generic HTTP_{n} for others
      const code: FederationClientErrorCode = status === 403 ? 'FORBIDDEN' : `HTTP_${status}`;
      throw new FederationClientError({
        status,
        code,
        message: `Peer ${peerId} returned ${status}: ${serverMessage}`,
        peerId,
      });
    }
    try {
      return JSON.parse(rawBody) as unknown;
    } catch (err) {
      throw new FederationClientError({
        code: 'INVALID_RESPONSE',
        message: `Peer ${peerId} returned non-JSON body`,
        peerId,
        cause: err,
      });
    }
  }
  /**
   * Parse and validate a response body against a Zod schema.
   *
   * For list/get, callers pass the result of `FederationListResponseSchema(z.unknown())`
   * so that the envelope structure is validated without requiring a concrete item schema
   * at the client level.  The generic `T` provides compile-time typing.
   *
   * Throws `FederationClientError({ code: 'INVALID_RESPONSE' })` on parse failure.
   */
  private parseWith<T>(peerId: string, body: unknown, schema: z.ZodTypeAny): T {
    const result = schema.safeParse(body);
    if (!result.success) {
      const issues = result.error.issues
        .map((e: z.ZodIssue) => `[${e.path.join('.') || 'root'}] ${e.message}`)
        .join('; ');
      throw new FederationClientError({
        code: 'INVALID_RESPONSE',
        message: `Peer ${peerId} returned invalid response shape: ${issues}`,
        peerId,
      });
    }
    return result.data as T;
  }
 }
--- a/apps/gateway/src/federation/client/index.ts
+++ b/apps/gateway/src/federation/client/index.ts
@@ -1,13 +0,0 @@
 /**
 * Federation client barrel — re-exports for FederationModule consumers.
 *
 * M3-09 (QuerySourceService) and future milestones should import from here,
 * not directly from the implementation file.
 */
 export {
  FederationClientService,
  FederationClientError,
  type FederationClientErrorCode,
  type FederationClientErrorOptions,
 } from './federation-client.service.js';
--- a/apps/gateway/src/federation/federation.module.ts
+++ b/apps/gateway/src/federation/federation.module.ts
@@ -5,25 +5,10 @@ import { EnrollmentController } from './enrollment.controller.js';
 import { EnrollmentService } from './enrollment.service.js';
 import { FederationController } from './federation.controller.js';
 import { GrantsService } from './grants.service.js';
 import { FederationClientService } from './client/index.js';
 import { FederationAuthGuard } from './server/index.js';
@Module({
  controllers: [EnrollmentController, FederationController],
-  providers: [
+  providers: [AdminGuard, CaService, EnrollmentService, GrantsService],
-    AdminGuard,
+  exports: [CaService, EnrollmentService, GrantsService],
    CaService,
    EnrollmentService,
    GrantsService,
    FederationClientService,
    FederationAuthGuard,
  ],
  exports: [
    CaService,
    EnrollmentService,
    GrantsService,
    FederationClientService,
    FederationAuthGuard,
  ],
 })
 export class FederationModule {}
--- a/apps/gateway/src/federation/grants.service.ts
+++ b/apps/gateway/src/federation/grants.service.ts
@@ -10,14 +10,12 @@
 */
 import { ConflictException, Inject, Injectable, NotFoundException } from '@nestjs/common';
-import { type Db, and, eq, federationGrants, federationPeers } from '@mosaicstack/db';
+import { type Db, and, eq, federationGrants } from '@mosaicstack/db';
 import { DB } from '../database/database.module.js';
 import { parseFederationScope } from './scope-schema.js';
 import type { CreateGrantDto, ListGrantsDto } from './grants.dto.js';
 export type Grant = typeof federationGrants.$inferSelect;
 export type Peer = typeof federationPeers.$inferSelect;
 export type GrantWithPeer = Grant & { peer: Peer };
@Injectable()
 export class GrantsService {
@@ -62,33 +60,6 @@ export class GrantsService {
    return grant;
  }
  /**
   * Fetch a single grant by ID, joined with its associated peer row.
   * Used by FederationAuthGuard to perform grant status + cert serial checks
   * in a single DB round-trip.
   *
   * Throws NotFoundException if the grant does not exist.
   * Throws NotFoundException if the associated peer row is missing (data integrity issue).
   */
  async getGrantWithPeer(id: string): Promise<GrantWithPeer> {
    const rows = await this.db
      .select()
      .from(federationGrants)
      .innerJoin(federationPeers, eq(federationGrants.peerId, federationPeers.id))
      .where(eq(federationGrants.id, id))
      .limit(1);
    const row = rows[0];
    if (!row) {
      throw new NotFoundException(`Grant ${id} not found`);
    }
    return {
      ...row.federation_grants,
      peer: row.federation_peers,
    };
  }
  /**
   * List grants with optional filters for peerId, subjectUserId, and status.
   */
--- a/apps/gateway/src/federation/oid.util.ts
+++ b/apps/gateway/src/federation/oid.util.ts
@@ -1,146 +0,0 @@
 /**
 * Shared OID extraction helpers for Mosaic federation certificates.
 *
 * Custom OID registry (PRD §6, docs/federation/SETUP.md):
 *   1.3.6.1.4.1.99999.1  — mosaic_grant_id
 *   1.3.6.1.4.1.99999.2  — mosaic_subject_user_id
 *
 * The encoding convention: each extension value is an OCTET STRING wrapping
 * an ASN.1 UTF8String TLV:
 *   0x0C (tag) + 1-byte length + UTF-8 bytes
 *
 * CaService encodes values this way via encodeUtf8String(), and this module
 * decodes them with the corresponding `.slice(2)` to skip tag + length byte.
 *
 * This module is intentionally pure — no NestJS, no DB, no network I/O.
 */
 import { X509Certificate } from '@peculiar/x509';
 // ---------------------------------------------------------------------------
 // OID constants
 // ---------------------------------------------------------------------------
 export const OID_MOSAIC_GRANT_ID = '1.3.6.1.4.1.99999.1';
 export const OID_MOSAIC_SUBJECT_USER_ID = '1.3.6.1.4.1.99999.2';
 // ---------------------------------------------------------------------------
 // Extraction result types
 // ---------------------------------------------------------------------------
 export interface MosaicOids {
  grantId: string;
  subjectUserId: string;
 }
 export type OidExtractionResult =
  | { ok: true; value: MosaicOids }
  | {
      ok: false;
      error: 'MISSING_GRANT_ID' | 'MISSING_SUBJECT_USER_ID' | 'PARSE_ERROR';
      detail?: string;
    };
 // ---------------------------------------------------------------------------
 // Helpers
 // ---------------------------------------------------------------------------
 const decoder = new TextDecoder();
 /**
 * Decode an extension value encoded as ASN.1 UTF8String TLV
 * (tag 0x0C + 1-byte length + UTF-8 bytes).
 * Validates tag, length byte, and buffer bounds before decoding.
 * Throws a descriptive Error on malformed input; caller wraps in try/catch.
 */
 function decodeUtf8StringTlv(value: ArrayBuffer): string {
  const bytes = new Uint8Array(value);
  // Need at least tag + length bytes
  if (bytes.length < 2) {
    throw new Error(`UTF8String TLV too short: expected at least 2 bytes, got ${bytes.length}`);
  }
  // Tag byte must be 0x0C (ASN.1 UTF8String)
  if (bytes[0] !== 0x0c) {
    throw new Error(
      `UTF8String TLV tag mismatch: expected 0x0C, got 0x${bytes[0]!.toString(16).toUpperCase()}`,
    );
  }
  // Only single-byte length form is supported (values 0–127); long form not needed
  // for OID strings of this length.
  const declaredLength = bytes[1]!;
  if (declaredLength > 127) {
    throw new Error(
      `UTF8String TLV uses long-form length (0x${declaredLength.toString(16).toUpperCase()}), which is not supported`,
    );
  }
  // Declared length must match actual remaining bytes
  if (declaredLength !== bytes.length - 2) {
    throw new Error(
      `UTF8String TLV length mismatch: declared ${declaredLength}, actual ${bytes.length - 2}`,
    );
  }
  // Skip: tag (1 byte) + length (1 byte)
  return decoder.decode(bytes.slice(2));
 }
 // ---------------------------------------------------------------------------
 // Public API
 // ---------------------------------------------------------------------------
 /**
 * Extract Mosaic custom OIDs (grantId, subjectUserId) from an X.509 certificate
 * already parsed via @peculiar/x509.
 *
 * Returns `{ ok: true, value: MosaicOids }` on success, or
 * `{ ok: false, error: <code>, detail? }` on any failure — never throws.
 */
 export function extractMosaicOids(cert: X509Certificate): OidExtractionResult {
  try {
    const grantIdExt = cert.getExtension(OID_MOSAIC_GRANT_ID);
    if (!grantIdExt) {
      return { ok: false, error: 'MISSING_GRANT_ID' };
    }
    const subjectUserIdExt = cert.getExtension(OID_MOSAIC_SUBJECT_USER_ID);
    if (!subjectUserIdExt) {
      return { ok: false, error: 'MISSING_SUBJECT_USER_ID' };
    }
    const grantId = decodeUtf8StringTlv(grantIdExt.value);
    const subjectUserId = decodeUtf8StringTlv(subjectUserIdExt.value);
    return {
      ok: true,
      value: { grantId, subjectUserId },
    };
  } catch (err) {
    return {
      ok: false,
      error: 'PARSE_ERROR',
      detail: err instanceof Error ? err.message : String(err),
    };
  }
 }
 /**
 * Parse a PEM-encoded certificate and extract Mosaic OIDs.
 * Returns an OidExtractionResult — never throws.
 */
 export function extractMosaicOidsFromPem(certPem: string): OidExtractionResult {
  let cert: X509Certificate;
  try {
    cert = new X509Certificate(certPem);
  } catch (err) {
    return {
      ok: false,
      error: 'PARSE_ERROR',
      detail: err instanceof Error ? err.message : String(err),
    };
  }
  return extractMosaicOids(cert);
 }
--- a/apps/gateway/src/federation/server/tests/federation-auth.guard.spec.ts
+++ b/apps/gateway/src/federation/server/tests/federation-auth.guard.spec.ts
@@ -1,521 +0,0 @@
 /**
 * Unit tests for FederationAuthGuard (FED-M3-03).
 *
 * Coverage:
 *  - Missing cert (no TLS socket / no getPeerCertificate) → 401
 *  - Cert parse failure (corrupt DER raw bytes) → 401
 *  - Missing grantId OID → 401
 *  - Missing subjectUserId OID → 401
 *  - Grant not found (GrantsService throws NotFoundException) → 403
 *  - Grant in `pending` status → 403
 *  - Grant in `revoked` status → 403
 *  - Grant in `expired` status → 403
 *  - Cert serial mismatch → 403
 *  - Happy path: active grant + matching cert serial → context attached, returns true
 */
 import 'reflect-metadata';
 import { describe, it, expect, vi, beforeEach } from 'vitest';
 import type { ExecutionContext } from '@nestjs/common';
 import { NotFoundException } from '@nestjs/common';
 import { FederationAuthGuard } from '../federation-auth.guard.js';
 import { makeMosaicIssuedCert } from '../../__tests__/helpers/test-cert.js';
 import type { GrantsService, GrantWithPeer } from '../../grants.service.js';
 // ---------------------------------------------------------------------------
 // Test constants
 // ---------------------------------------------------------------------------
 const GRANT_ID = 'a1111111-1111-1111-1111-111111111111';
 const USER_ID = 'b2222222-2222-2222-2222-222222222222';
 const PEER_ID = 'c3333333-3333-3333-3333-333333333333';
 // Node.js TLS serialNumber is uppercase hex (no colons)
 const CERT_SERIAL_HEX = '01';
 const VALID_SCOPE = { resources: ['tasks'], max_rows_per_query: 100 };
 // ---------------------------------------------------------------------------
 // Mock builders
 // ---------------------------------------------------------------------------
 /**
 * Build a minimal GrantWithPeer-shaped mock.
 */
 function makeGrantWithPeer(overrides: Partial<GrantWithPeer> = {}): GrantWithPeer {
  return {
    id: GRANT_ID,
    peerId: PEER_ID,
    subjectUserId: USER_ID,
    scope: VALID_SCOPE,
    status: 'active',
    expiresAt: null,
    createdAt: new Date('2026-01-01T00:00:00Z'),
    revokedAt: null,
    revokedReason: null,
    peer: {
      id: PEER_ID,
      commonName: 'test-peer',
      displayName: 'Test Peer',
      certPem: '',
      certSerial: CERT_SERIAL_HEX,
      certNotAfter: new Date(Date.now() + 86_400_000),
      clientKeyPem: null,
      state: 'active',
      endpointUrl: null,
      lastSeenAt: null,
      createdAt: new Date('2026-01-01T00:00:00Z'),
      revokedAt: null,
    },
    ...overrides,
  };
 }
 /**
 * Build a mock ExecutionContext with a pre-built TLS peer certificate.
 *
 * `certPem` — PEM string to present as the raw DER cert (converted to Buffer).
 *             Pass null to simulate "no cert presented".
 * `certSerialHex` — serialNumber string returned by the TLS socket.
 *                   Node.js returns uppercase hex.
 * `hasTlsSocket` — if false, raw.socket has no getPeerCertificate (plain HTTP).
 */
 function makeContext(opts: {
  certPem: string | null;
  certSerialHex?: string;
  hasTlsSocket?: boolean;
 }): {
  ctx: ExecutionContext;
  statusMock: ReturnType<typeof vi.fn>;
  sendMock: ReturnType<typeof vi.fn>;
 } {
  const { certPem, certSerialHex = CERT_SERIAL_HEX, hasTlsSocket = true } = opts;
  // Build peerCert object that Node.js TLS socket.getPeerCertificate() returns
  let peerCert: Record<string, unknown>;
  if (certPem === null) {
    // Simulate no cert: Node.js returns object with empty string fields
    peerCert = { raw: null, serialNumber: '' };
  } else {
    // Convert PEM to DER Buffer (strip headers + base64 decode)
    const b64 = certPem
      .replace(/-----BEGIN CERTIFICATE-----/, '')
      .replace(/-----END CERTIFICATE-----/, '')
      .replace(/\s+/g, '');
    const raw = Buffer.from(b64, 'base64');
    peerCert = { raw, serialNumber: certSerialHex };
  }
  const getPeerCertificate = vi.fn().mockReturnValue(peerCert);
  const socket = hasTlsSocket ? { getPeerCertificate } : {}; // No getPeerCertificate → non-TLS
  // Fastify reply mocks
  const sendMock = vi.fn().mockReturnValue(undefined);
  const headerMock = vi.fn().mockReturnValue({ send: sendMock });
  const statusMock = vi.fn().mockReturnValue({ header: headerMock });
  const request = {
    raw: {
      socket,
    },
  };
  const reply = {
    status: statusMock,
  };
  const ctx = {
    switchToHttp: () => ({
      getRequest: () => request,
      getResponse: () => reply,
    }),
  } as unknown as ExecutionContext;
  return { ctx, statusMock, sendMock };
 }
 /**
 * Build a mock GrantsService.
 */
 function makeGrantsService(
  overrides: Partial<Pick<GrantsService, 'getGrantWithPeer'>> = {},
 ): GrantsService {
  return {
    getGrantWithPeer: vi.fn().mockResolvedValue(makeGrantWithPeer()),
    ...overrides,
  } as unknown as GrantsService;
 }
 // ---------------------------------------------------------------------------
 // Test suite
 // ---------------------------------------------------------------------------
 describe('FederationAuthGuard', () => {
  let certPem: string;
  beforeEach(async () => {
    // Generate a real Mosaic-issued cert with the standard OIDs
    certPem = await makeMosaicIssuedCert({ grantId: GRANT_ID, subjectUserId: USER_ID });
  });
  // ── 401: No TLS socket ────────────────────────────────────────────────────
  it('returns 401 when there is no TLS socket (plain HTTP connection)', async () => {
    const { ctx, statusMock, sendMock } = makeContext({
      certPem: certPem,
      hasTlsSocket: false,
    });
    const guard = new FederationAuthGuard(makeGrantsService());
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(401);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'unauthorized', message: expect.any(String) }),
      }),
    );
  });
  // ── 401: Cert not presented ───────────────────────────────────────────────
  it('returns 401 when the peer did not present a certificate', async () => {
    const { ctx, statusMock, sendMock } = makeContext({ certPem: null });
    const guard = new FederationAuthGuard(makeGrantsService());
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(401);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'unauthorized', message: expect.any(String) }),
      }),
    );
  });
  // ── 401: Cert parse failure ───────────────────────────────────────────────
  it('returns 401 when the certificate DER bytes are corrupt', async () => {
    // Build context with a cert that has garbage DER bytes
    const corruptPem = '-----BEGIN CERTIFICATE-----\naW52YWxpZA==\n-----END CERTIFICATE-----';
    const { ctx, statusMock, sendMock } = makeContext({ certPem: corruptPem });
    const guard = new FederationAuthGuard(makeGrantsService());
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(401);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'unauthorized', message: expect.any(String) }),
      }),
    );
  });
  // ── 401: Missing grantId OID ─────────────────────────────────────────────
  it('returns 401 when the cert is missing the grantId OID', async () => {
    // makeSelfSignedCert produces a cert without any Mosaic OIDs
    const { makeSelfSignedCert } = await import('../../__tests__/helpers/test-cert.js');
    const plainCert = await makeSelfSignedCert();
    const { ctx, statusMock, sendMock } = makeContext({ certPem: plainCert });
    const guard = new FederationAuthGuard(makeGrantsService());
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(401);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'unauthorized', message: expect.any(String) }),
      }),
    );
  });
  // ── 401: Missing subjectUserId OID ───────────────────────────────────────
  it('returns 401 when the cert has grantId OID but is missing subjectUserId OID', async () => {
    // Build a cert with only the grantId OID by importing cert generator internals
    const { webcrypto } = await import('node:crypto');
    const {
      X509CertificateGenerator,
      Extension,
      KeyUsagesExtension,
      KeyUsageFlags,
      BasicConstraintsExtension,
      cryptoProvider,
    } = await import('@peculiar/x509');
    cryptoProvider.set(webcrypto as unknown as Parameters<typeof cryptoProvider.set>[0]);
    const alg = { name: 'ECDSA', namedCurve: 'P-256', hash: 'SHA-256' } as const;
    const keys = await webcrypto.subtle.generateKey(alg, false, ['sign', 'verify']);
    const now = new Date();
    const tomorrow = new Date(now.getTime() + 86_400_000);
    // Encode grantId only — missing subjectUserId extension
    const utf8 = new TextEncoder().encode(GRANT_ID);
    const encoded = new Uint8Array(2 + utf8.length);
    encoded[0] = 0x0c;
    encoded[1] = utf8.length;
    encoded.set(utf8, 2);
    const cert = await X509CertificateGenerator.createSelfSigned({
      serialNumber: '01',
      name: 'CN=partial-oid-test',
      notBefore: now,
      notAfter: tomorrow,
      signingAlgorithm: alg,
      keys,
      extensions: [
        new BasicConstraintsExtension(false),
        new KeyUsagesExtension(KeyUsageFlags.digitalSignature),
        new Extension('1.3.6.1.4.1.99999.1', false, encoded), // grantId only
      ],
    });
    const { ctx, statusMock, sendMock } = makeContext({ certPem: cert.toString('pem') });
    const guard = new FederationAuthGuard(makeGrantsService());
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(401);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'unauthorized', message: expect.any(String) }),
      }),
    );
  });
  // ── 403: Grant not found ─────────────────────────────────────────────────
  it('returns 403 when the grantId from the cert does not exist in DB', async () => {
    const grantsService = makeGrantsService({
      getGrantWithPeer: vi
        .fn()
        .mockRejectedValue(new NotFoundException(`Grant ${GRANT_ID} not found`)),
    });
    const { ctx, statusMock, sendMock } = makeContext({ certPem });
    const guard = new FederationAuthGuard(grantsService);
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(403);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'forbidden', message: 'Federation access denied' }),
      }),
    );
  });
  // ── 403: Grant in `pending` status ───────────────────────────────────────
  it('returns 403 when the grant is in pending status', async () => {
    const grantsService = makeGrantsService({
      getGrantWithPeer: vi.fn().mockResolvedValue(makeGrantWithPeer({ status: 'pending' })),
    });
    const { ctx, statusMock, sendMock } = makeContext({ certPem });
    const guard = new FederationAuthGuard(grantsService);
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(403);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'forbidden', message: 'Federation access denied' }),
      }),
    );
  });
  // ── 403: Grant in `revoked` status ───────────────────────────────────────
  it('returns 403 when the grant is in revoked status', async () => {
    const grantsService = makeGrantsService({
      getGrantWithPeer: vi
        .fn()
        .mockResolvedValue(makeGrantWithPeer({ status: 'revoked', revokedAt: new Date() })),
    });
    const { ctx, statusMock, sendMock } = makeContext({ certPem });
    const guard = new FederationAuthGuard(grantsService);
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(403);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'forbidden', message: 'Federation access denied' }),
      }),
    );
  });
  // ── 403: Grant in `expired` status ───────────────────────────────────────
  it('returns 403 when the grant is in expired status', async () => {
    const grantsService = makeGrantsService({
      getGrantWithPeer: vi.fn().mockResolvedValue(makeGrantWithPeer({ status: 'expired' })),
    });
    const { ctx, statusMock, sendMock } = makeContext({ certPem });
    const guard = new FederationAuthGuard(grantsService);
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(403);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'forbidden', message: 'Federation access denied' }),
      }),
    );
  });
  // ── 403: Cert serial mismatch ─────────────────────────────────────────────
  it('returns 403 when the cert serial does not match the registered peer cert serial', async () => {
    // Return a grant whose peer has a different stored serial
    const grantsService = makeGrantsService({
      getGrantWithPeer: vi.fn().mockResolvedValue(
        makeGrantWithPeer({
          peer: {
            id: PEER_ID,
            commonName: 'test-peer',
            displayName: 'Test Peer',
            certPem: '',
            certSerial: 'DEADBEEF', // different from CERT_SERIAL_HEX='01'
            certNotAfter: new Date(Date.now() + 86_400_000),
            clientKeyPem: null,
            state: 'active',
            endpointUrl: null,
            lastSeenAt: null,
            createdAt: new Date('2026-01-01T00:00:00Z'),
            revokedAt: null,
          },
        }),
      ),
    });
    // Context presents cert with serial '01' but DB has 'DEADBEEF'
    const { ctx, statusMock, sendMock } = makeContext({ certPem, certSerialHex: '01' });
    const guard = new FederationAuthGuard(grantsService);
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(403);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'forbidden', message: 'Federation access denied' }),
      }),
    );
  });
  // ── 403: subjectUserId cert/DB mismatch (CRIT-1 regression test) ─────────
  it('returns 403 when the cert subjectUserId does not match the DB grant subjectUserId', async () => {
    // Build a cert that claims an attacker's subjectUserId
    const attackerSubjectUserId = 'attacker-user-id';
    const attackerCertPem = await makeMosaicIssuedCert({
      grantId: GRANT_ID,
      subjectUserId: attackerSubjectUserId,
    });
    // DB returns a grant with the legitimate USER_ID
    const grantsService = makeGrantsService({
      getGrantWithPeer: vi.fn().mockResolvedValue(makeGrantWithPeer({ subjectUserId: USER_ID })),
    });
    // Cert presents attacker-user-id but DB has USER_ID — should be rejected
    const { ctx, statusMock, sendMock } = makeContext({
      certPem: attackerCertPem,
      certSerialHex: CERT_SERIAL_HEX,
    });
    const guard = new FederationAuthGuard(grantsService);
    const result = await guard.canActivate(ctx);
    expect(result).toBe(false);
    expect(statusMock).toHaveBeenCalledWith(403);
    expect(sendMock).toHaveBeenCalledWith(
      expect.objectContaining({
        error: expect.objectContaining({ code: 'forbidden', message: 'Federation access denied' }),
      }),
    );
  });
  // ── Happy path ────────────────────────────────────────────────────────────
  it('returns true and attaches federationContext on happy path', async () => {
    const grant = makeGrantWithPeer({
      status: 'active',
      peer: {
        id: PEER_ID,
        commonName: 'test-peer',
        displayName: 'Test Peer',
        certPem: '',
        certSerial: CERT_SERIAL_HEX,
        certNotAfter: new Date(Date.now() + 86_400_000),
        clientKeyPem: null,
        state: 'active',
        endpointUrl: null,
        lastSeenAt: null,
        createdAt: new Date('2026-01-01T00:00:00Z'),
        revokedAt: null,
      },
    });
    const grantsService = makeGrantsService({
      getGrantWithPeer: vi.fn().mockResolvedValue(grant),
    });
    // Build context manually to capture what gets set on request.federationContext
    const b64 = certPem
      .replace(/-----BEGIN CERTIFICATE-----/, '')
      .replace(/-----END CERTIFICATE-----/, '')
      .replace(/\s+/g, '');
    const raw = Buffer.from(b64, 'base64');
    const peerCert = { raw, serialNumber: CERT_SERIAL_HEX };
    const sendMock = vi.fn().mockReturnValue(undefined);
    const headerMock = vi.fn().mockReturnValue({ send: sendMock });
    const statusMock = vi.fn().mockReturnValue({ header: headerMock });
    const request: Record<string, unknown> = {
      raw: {
        socket: { getPeerCertificate: vi.fn().mockReturnValue(peerCert) },
      },
    };
    const reply = { status: statusMock };
    const ctx = {
      switchToHttp: () => ({
        getRequest: () => request,
        getResponse: () => reply,
      }),
    } as unknown as ExecutionContext;
    const guard = new FederationAuthGuard(grantsService);
    const result = await guard.canActivate(ctx);
    expect(result).toBe(true);
    expect(statusMock).not.toHaveBeenCalled();
    // Verify the context was attached correctly
    expect(request['federationContext']).toEqual({
      grantId: GRANT_ID,
      subjectUserId: USER_ID,
      peerId: PEER_ID,
      scope: VALID_SCOPE,
    });
  });
 });
--- a/apps/gateway/src/federation/server/federation-auth.guard.ts
+++ b/apps/gateway/src/federation/server/federation-auth.guard.ts
@@ -1,212 +0,0 @@
 /**
 * FederationAuthGuard — NestJS CanActivate guard for inbound federation requests.
 *
 * Validates the mTLS client certificate presented by a peer gateway, extracts
 * custom OIDs to identify the grant + subject user, loads the grant from DB,
 * asserts it is active, and verifies the cert serial against the registered peer
 * cert serial as a defense-in-depth measure.
 *
 * On success, attaches `request.federationContext` for downstream verb controllers.
 * On failure, responds with the federation wire-format error envelope (not raw
 * NestJS exception JSON) to match the federation protocol contract.
 *
 * ## Cert-serial check decision
 * The guard validates that the inbound client cert's serial number matches the
 * `certSerial` stored on the associated `federation_peers` row. This is a
 * defense-in-depth measure: even if the mTLS handshake is compromised at the
 * transport layer (e.g. misconfigured TLS terminator that forwards arbitrary
 * client certs), an attacker cannot replay a cert with a different serial than
 * what was registered during enrollment. This check is NOT loosened because:
 *  1. It is O(1) — no additional DB round-trip (peerId is on the grant row,
 *     so we join to federationPeers in the same query).
 *  2. Cert renewal MUST update the stored serial — enforced by M6 scheduler.
 *  3. The OID-only path (without serial check) would allow any cert from the
 *     same CA bearing the same grantId OID to succeed after cert compromise.
 *
 * ## FastifyRequest typing path
 * NestJS + Fastify wraps the raw Node.js IncomingMessage in a FastifyRequest.
 * The underlying TLS socket is accessed via `request.raw.socket`, which is a
 * `tls.TLSSocket` when the server is listening on HTTPS. In development/test
 * the gateway may run over plain HTTP, in which case `getPeerCertificate` is
 * not available. The guard safely handles both cases by checking for the
 * method's existence before calling it.
 *
 * Note: The guard reads the peer certificate from the *already-completed*
 * TLS handshake via `socket.getPeerCertificate(detailed=true)`. This relies
 * on the server being configured with `requestCert: true` at the TLS level
 * so Fastify/Node.js requests the client cert during the handshake.
 * The guard does NOT verify the cert chain itself — that is handled by the
 * TLS layer (Node.js `rejectUnauthorized: true` with the CA cert pinned).
 */
 import {
  type CanActivate,
  type ExecutionContext,
  Inject,
  Injectable,
  Logger,
 } from '@nestjs/common';
 import type { FastifyReply, FastifyRequest } from 'fastify';
 import * as tls from 'node:tls';
 import { X509Certificate } from '@peculiar/x509';
 import { FederationForbiddenError, FederationUnauthorizedError } from '@mosaicstack/types';
 import { extractMosaicOids } from '../oid.util.js';
 import { GrantsService } from '../grants.service.js';
 import type { FederationContext } from './federation-context.js';
 import './federation-context.js'; // side-effect import: applies FastifyRequest module augmentation
 // ---------------------------------------------------------------------------
 // Internal helpers
 // ---------------------------------------------------------------------------
 /**
 * Send a federation wire-format error response directly on the Fastify reply.
 * Returns false — callers return this value from canActivate.
 */
 function sendFederationError(
  reply: FastifyReply,
  error: FederationUnauthorizedError | FederationForbiddenError,
 ): boolean {
  const statusCode = error.code === 'unauthorized' ? 401 : 403;
  void reply.status(statusCode).header('content-type', 'application/json').send(error.toEnvelope());
  return false;
 }
 // ---------------------------------------------------------------------------
 // Guard
 // ---------------------------------------------------------------------------
@Injectable()
 export class FederationAuthGuard implements CanActivate {
  private readonly logger = new Logger(FederationAuthGuard.name);
  constructor(@Inject(GrantsService) private readonly grantsService: GrantsService) {}
  async canActivate(context: ExecutionContext): Promise<boolean> {
    const http = context.switchToHttp();
    const request = http.getRequest<FastifyRequest>();
    const reply = http.getResponse<FastifyReply>();
    // ── Step 1: Extract peer certificate from TLS socket ────────────────────
    const rawSocket = request.raw.socket;
    // Check TLS socket: getPeerCertificate is only available on TLS connections.
    if (
      !rawSocket ||
      typeof (rawSocket as Partial<tls.TLSSocket>).getPeerCertificate !== 'function'
    ) {
      this.logger.warn('No TLS socket — client cert unavailable (non-mTLS connection)');
      return sendFederationError(
        reply,
        new FederationUnauthorizedError('Client certificate required'),
      );
    }
    const tlsSocket = rawSocket as tls.TLSSocket;
    const peerCert = tlsSocket.getPeerCertificate(true);
    // Node.js returns an object with empty string fields when no cert was presented.
    if (!peerCert || !peerCert.raw) {
      this.logger.warn('Peer certificate not presented (mTLS handshake did not supply cert)');
      return sendFederationError(
        reply,
        new FederationUnauthorizedError('Client certificate required'),
      );
    }
    // ── Step 2: Parse the DER-encoded certificate via @peculiar/x509 ────────
    let cert: X509Certificate;
    try {
      // peerCert.raw is a Buffer containing the DER-encoded cert
      cert = new X509Certificate(peerCert.raw);
    } catch (err) {
      this.logger.warn(
        `Failed to parse peer certificate: ${err instanceof Error ? err.message : String(err)}`,
      );
      return sendFederationError(
        reply,
        new FederationUnauthorizedError('Client certificate could not be parsed'),
      );
    }
    // ── Step 3: Extract Mosaic custom OIDs ──────────────────────────────────
    const oidResult = extractMosaicOids(cert);
    if (!oidResult.ok) {
      const message =
        oidResult.error === 'MISSING_GRANT_ID'
          ? 'Client certificate is missing required OID: mosaic_grant_id (1.3.6.1.4.1.99999.1)'
          : oidResult.error === 'MISSING_SUBJECT_USER_ID'
            ? 'Client certificate is missing required OID: mosaic_subject_user_id (1.3.6.1.4.1.99999.2)'
            : `Client certificate OID extraction failed: ${oidResult.detail ?? 'unknown error'}`;
      this.logger.warn(`OID extraction failure [${oidResult.error}]: ${message}`);
      return sendFederationError(reply, new FederationUnauthorizedError(message));
    }
    const { grantId, subjectUserId } = oidResult.value;
    // ── Step 4: Load grant from DB ───────────────────────────────────────────
    let grant: Awaited<ReturnType<GrantsService['getGrantWithPeer']>>;
    try {
      grant = await this.grantsService.getGrantWithPeer(grantId);
    } catch {
      // getGrantWithPeer throws NotFoundException when not found
      this.logger.warn(`Grant not found: ${grantId}`);
      return sendFederationError(reply, new FederationForbiddenError('Federation access denied'));
    }
    // ── Step 5: Assert grant is active ──────────────────────────────────────
    if (grant.status !== 'active') {
      this.logger.warn(`Grant ${grantId} is not active — status=${grant.status}`);
      return sendFederationError(reply, new FederationForbiddenError('Federation access denied'));
    }
    // ── Step 5b: Validate cert-extracted subjectUserId against DB (CRIT-1) ──
    // The cert claim is untrusted input; the DB row is authoritative.
    if (subjectUserId !== grant.subjectUserId) {
      this.logger.warn(`subjectUserId mismatch for grant ${grantId}`);
      return sendFederationError(reply, new FederationForbiddenError('Federation access denied'));
    }
    // ── Step 6: Defense-in-depth — cert serial must match registered peer ───
    // The serial number from Node.js TLS is upper-case hex without colons.
    // The @peculiar/x509 serialNumber is decimal. We compare using the native
    // Node.js crypto cert serial which is uppercase hex, matching DB storage.
    // Both are derived from the peerCert.serialNumber Node.js provides.
    const inboundSerial: string = peerCert.serialNumber ?? '';
    if (!grant.peer.certSerial) {
      // Peer row exists but has no stored serial — something is wrong with enrollment
      this.logger.error(`Peer ${grant.peerId} has no stored certSerial — enrollment incomplete`);
      return sendFederationError(reply, new FederationForbiddenError('Federation access denied'));
    }
    // Normalize both to uppercase for comparison (Node.js serialNumber is
    // already uppercase hex; DB value was stored from extractSerial() which
    // returns crypto.X509Certificate.serialNumber — also uppercase hex).
    if (inboundSerial.toUpperCase() !== grant.peer.certSerial.toUpperCase()) {
      this.logger.warn(
        `Cert serial mismatch for grant ${grantId}: ` +
          `inbound=${inboundSerial} registered=${grant.peer.certSerial}`,
      );
      return sendFederationError(reply, new FederationForbiddenError('Federation access denied'));
    }
    // ── Step 7: Attach FederationContext to request ──────────────────────────
    // Use grant.subjectUserId from DB (authoritative) — not the cert-extracted value.
    const federationContext: FederationContext = {
      grantId,
      subjectUserId: grant.subjectUserId,
      peerId: grant.peerId,
      scope: grant.scope as Record<string, unknown>,
    };
    request.federationContext = federationContext;
    this.logger.debug(
      `Federation auth OK — grantId=${grantId} peerId=${grant.peerId} subjectUserId=${grant.subjectUserId}`,
    );
    return true;
  }
 }
--- a/apps/gateway/src/federation/server/federation-context.ts
+++ b/apps/gateway/src/federation/server/federation-context.ts
@@ -1,39 +0,0 @@
 /**
 * FederationContext — attached to inbound federation requests after successful
 * mTLS + grant validation by FederationAuthGuard.
 *
 * Downstream verb controllers access this via `request.federationContext`.
 */
 /**
 * Augment FastifyRequest so TypeScript knows about the federation context
 * property that FederationAuthGuard attaches on success.
 */
 declare module 'fastify' {
  interface FastifyRequest {
    federationContext?: FederationContext;
  }
 }
 /**
 * Typed context object attached to the request by FederationAuthGuard.
 * Carries all data extracted from the mTLS cert + grant DB row needed
 * by downstream federation verb handlers.
 */
 export interface FederationContext {
  /** The federation grant ID extracted from OID 1.3.6.1.4.1.99999.1 */
  grantId: string;
  /** The local subject user whose data is accessible under this grant */
  subjectUserId: string;
  /** The peer gateway ID (from the grant's peerId FK) */
  peerId: string;
  /**
   * Grant scope — determines which resources the peer may query.
   * Typed as Record<string, unknown> because the full scope schema lives in
   * scope-schema.ts; downstream handlers should narrow via parseFederationScope.
   */
  scope: Record<string, unknown>;
 }
--- a/apps/gateway/src/federation/server/index.ts
+++ b/apps/gateway/src/federation/server/index.ts
@@ -1,13 +0,0 @@
 /**
 * Federation server-side barrel — inbound request handling.
 *
 * Exports the mTLS auth guard and the FederationContext interface
 * for use by verb controllers (M3-05/06/07).
 *
 * Usage:
 *   import { FederationAuthGuard } from './server/index.js';
 *   @UseGuards(FederationAuthGuard)
 */
 export { FederationAuthGuard } from './federation-auth.guard.js';
 export type { FederationContext } from './federation-context.js';
--- a/docs/scratchpads/mvp-20260312.md
+++ b/docs/scratchpads/mvp-20260312.md
@@ -653,9 +653,3 @@ Independent security review surfaced three high-impact and four medium findings;
 2. After merge, kickoff M3-01 (DTOs) on `feat/federation-m3-types` with sonnet subagent in worktree
 3. Once M3-01 lands, fan out: M3-02 (harness) || M3-03 (AuthGuard) → M3-04 (ScopeService) || M3-08 (FederationClient)
 4. Re-converge at M3-10 (Integration) → M3-11 (E2E)
 ### Session 24 — 2026-05-22 — Gitea PR metadata wrapper hardening
 - Fixed `packages/mosaic/framework/tools/git/pr-metadata.sh` Gitea path to fail closed on non-2xx API responses instead of normalizing API error JSON into null/empty PR metadata.
 - Added token fallback behavior: try explicit `GITEA_TOKEN`, Mosaic credential-loader token, then matching `~/.git-credentials` HTTPS token; this handles stale host-scoped credential-loader entries while preserving existing credential sources.
 - Confirmed real U-Connect Gitea PRs #1905 and #1908 now return `number`, `headRefName`, `baseRefName`, `state`, `author`, `url`, and `mergeable` correctly, restoring `pr-merge.sh` base branch detection.
--- a/packages/db/package.json
+++ b/packages/db/package.json
@@ -42,7 +42,6 @@
    "access": "public"
  },
  "files": [
-    "dist",
+    "dist"
    "drizzle"
  ]
 }
--- a/packages/db/src/client-pglite.ts
+++ b/packages/db/src/client-pglite.ts
@@ -1,12 +1,10 @@
 import { PGlite } from '@electric-sql/pglite';
 import { vector } from '@electric-sql/pglite/vector';
 import { drizzle } from 'drizzle-orm/pglite';
 import * as schema from './schema.js';
 import type { DbHandle } from './client.js';
 export function createPgliteDb(dataDir: string): DbHandle {
-  // pgvector extension is required by migration 0001 (insights.embedding column).
+  const client = new PGlite(dataDir);
  const client = new PGlite(dataDir, { extensions: { vector } });
  const db = drizzle(client, { schema });
  return {
    db: db as unknown as DbHandle['db'],
--- a/packages/db/src/index.ts
+++ b/packages/db/src/index.ts
@@ -1,6 +1,6 @@
 export { createDb, type Db, type DbHandle } from './client.js';
 export { createPgliteDb } from './client-pglite.js';
-export { runMigrations, runPgliteMigrations } from './migrate.js';
+export { runMigrations } from './migrate.js';
 export * from './schema.js';
 export * from './federation.js';
 export {
--- a/packages/db/src/migrate.test.ts
+++ b/packages/db/src/migrate.test.ts
@@ -1,70 +0,0 @@
 import { mkdtempSync, rmSync } from 'node:fs';
 import { tmpdir } from 'node:os';
 import { join } from 'node:path';
 import { sql } from 'drizzle-orm';
 import { afterEach, beforeEach, describe, expect, it } from 'vitest';
 import { createPgliteDb } from './client-pglite.js';
 import { runPgliteMigrations } from './migrate.js';
 import type { DbHandle } from './client.js';
 interface PgliteExec {
  exec(query: string): Promise<unknown>;
 }
 describe('runPgliteMigrations', () => {
  let dataDir: string;
  let handle: DbHandle;
  beforeEach(() => {
    dataDir = mkdtempSync(join(tmpdir(), 'mosaic-db-migrate-test-'));
    handle = createPgliteDb(dataDir);
  });
  afterEach(async () => {
    await handle.close();
    rmSync(dataDir, { recursive: true, force: true });
  });
  it('creates the BetterAuth tables required by the gateway', async () => {
    await runPgliteMigrations(handle);
    const result = (await handle.db.execute(sql`
      SELECT table_name FROM information_schema.tables
      WHERE table_schema = 'public'
      ORDER BY table_name
    `)) as unknown as { rows: Array<{ table_name: string }> };
    const tables = result.rows.map((r) => r.table_name);
    // Auth tables — required for sign-in / bootstrap to function.
    expect(tables).toContain('users');
    expect(tables).toContain('sessions');
    expect(tables).toContain('accounts');
    expect(tables).toContain('verifications');
    // Schema sanity check — admin token table consumed by mosaic gateway config.
    expect(tables).toContain('admin_tokens');
  });
  it('is idempotent — running twice does not error', async () => {
    await runPgliteMigrations(handle);
    await expect(runPgliteMigrations(handle)).resolves.toBeUndefined();
  });
  it('surfaces statement-level error context on failure and leaves no ledger row', async () => {
    // Pre-create a `users` table that conflicts with migration 0000's CREATE TABLE,
    // forcing it to fail without IF NOT EXISTS.
    const client = (handle.db as unknown as { $client: PgliteExec }).$client;
    await client.exec('CREATE TABLE users (sentinel text)');
    await expect(runPgliteMigrations(handle)).rejects.toThrow(
      /migration hash=[a-f0-9]+ statement #\d+ failed/,
    );
    // Ledger should be empty — partial application must not pretend to be complete.
    const ledger = (await handle.db.execute(
      sql`SELECT count(*)::int AS count FROM drizzle.__drizzle_migrations`,
    )) as unknown as { rows: Array<{ count: number }> };
    expect(ledger.rows[0]?.count).toBe(0);
  });
 });
--- a/packages/db/src/migrate.ts
+++ b/packages/db/src/migrate.ts
@@ -1,109 +1,18 @@
 import { dirname, resolve } from 'node:path';
 import { fileURLToPath } from 'node:url';
-import { sql } from 'drizzle-orm';
+import { drizzle } from 'drizzle-orm/postgres-js';
-import { drizzle as drizzlePostgres } from 'drizzle-orm/postgres-js';
+import { migrate } from 'drizzle-orm/postgres-js/migrator';
 import { migrate as migratePostgres } from 'drizzle-orm/postgres-js/migrator';
 import { readMigrationFiles } from 'drizzle-orm/migrator';
 import postgres from 'postgres';
 import { DEFAULT_DATABASE_URL } from './defaults.js';
 import type { DbHandle } from './client.js';
 interface PgliteExecutor {
  exec(query: string): Promise<unknown>;
 }
 interface ExecuteRows<T> {
  rows: T[];
 }
 function migrationsFolder(): string {
  const here = dirname(fileURLToPath(import.meta.url));
  return resolve(here, '../drizzle');
 }
 export async function runMigrations(url?: string): Promise<void> {
  const connectionString = url ?? process.env['DATABASE_URL'] ?? DEFAULT_DATABASE_URL;
-  const sqlClient = postgres(connectionString, { max: 1 });
+  const sql = postgres(connectionString, { max: 1 });
-  const db = drizzlePostgres(sqlClient);
+  const db = drizzle(sql);
  const __dirname = dirname(fileURLToPath(import.meta.url));
  try {
-    // TODO: postgres-tier first-install also fails because (a) Drizzle wraps every
+    await migrate(db, { migrationsFolder: resolve(__dirname, '../drizzle') });
    // migration in one transaction (breaks 0009's ALTER TYPE ADD VALUE → SET DEFAULT
    // sequence) and (b) drizzle/meta/_journal.json has 0009 ordered before 0008,
    // which the postgres-js migrator skips by `created_at < folderMillis`. The
    // PGlite path below sidesteps both. A follow-up should either share the
    // per-statement loop (see runPgliteMigrations) or fix the journal ordering.
    await migratePostgres(db, { migrationsFolder: migrationsFolder() });
  } finally {
-    await sqlClient.end();
+    await sql.end();
  }
 }
 // Apply Drizzle migrations against an embedded PGlite database.
 //
 // We don't reuse drizzle's pglite migrator because it wraps ALL migrations in
 // one outer transaction, which breaks Postgres' `check_safe_enum_use` rule —
 // e.g. migration 0009 does `ALTER TYPE ADD VALUE 'pending'` then references
 // `'pending'` as a default in the same tx. PGlite's `exec()` runs each
 // statement under the Simple Query protocol, autocommitting between them.
 //
 // We still write to the standard `drizzle.__drizzle_migrations` ledger so the
 // result is interoperable with `runMigrations()` on a postgres-backed deploy
 // (modulo the journal-ordering bug noted above).
 //
 // We skip-by-hash rather than skip-by-folderMillis (which is what Drizzle's
 // postgres-js migrator does). That's deliberate — out-of-order timestamps in
 // `_journal.json` won't silently drop migrations.
 //
 // Failure model: each statement autocommits, and the ledger row is written
 // only after all statements in a migration succeed. A crash mid-migration
 // leaves the prefix applied with no ledger entry, so the next boot will
 // replay those statements and fail loudly on "already exists". Recovery:
 // drop the partially-applied objects, or insert the migration's hash into
 // `drizzle.__drizzle_migrations` manually. The error log identifies which
 // statement of which migration was the culprit.
 export async function runPgliteMigrations(handle: DbHandle): Promise<void> {
  const client = (handle.db as unknown as { $client?: PgliteExecutor }).$client;
  if (!client || typeof client.exec !== 'function') {
    throw new Error('runPgliteMigrations: handle.db is not backed by a PGlite client');
  }
  await client.exec('CREATE SCHEMA IF NOT EXISTS drizzle');
  await client.exec(`
    CREATE TABLE IF NOT EXISTS drizzle.__drizzle_migrations (
      id SERIAL PRIMARY KEY,
      hash text NOT NULL,
      created_at bigint
    )
  `);
  const appliedRows = (await handle.db.execute(
    sql`SELECT hash FROM drizzle.__drizzle_migrations`,
  )) as unknown as ExecuteRows<{ hash: string }>;
  const applied = new Set(appliedRows.rows.map((r) => r.hash));
  const migrations = readMigrationFiles({ migrationsFolder: migrationsFolder() });
  for (const migration of migrations) {
    if (applied.has(migration.hash)) continue;
    // Run each statement-breakpoint chunk in its own exec() call so PGlite
    // commits between statements — this is what lets `ALTER TYPE ADD VALUE`
    // become visible before a subsequent statement references the new value.
    for (const [stmtIdx, stmt] of migration.sql.entries()) {
      const trimmed = stmt.trim();
      if (!trimmed) continue;
      try {
        await client.exec(trimmed);
      } catch (err) {
        const cause = err instanceof Error ? err.message : String(err);
        throw new Error(
          `runPgliteMigrations: migration hash=${migration.hash} statement #${stmtIdx} failed: ${cause}\n` +
            `Statement: ${trimmed.slice(0, 200)}${trimmed.length > 200 ? '…' : ''}`,
          { cause: err },
        );
      }
    }
    await handle.db.execute(
      sql`INSERT INTO drizzle.__drizzle_migrations (hash, created_at) VALUES (${migration.hash}, ${migration.folderMillis})`,
    );
  }
 }
--- a/packages/mosaic/framework/tools/git/pr-metadata.sh
+++ b/packages/mosaic/framework/tools/git/pr-metadata.sh
@@ -69,81 +69,32 @@ elif [[ "$PLATFORM" == "gitea" ]]; then
    API_URL="https://${HOST}/api/v1/repos/${OWNER}/${REPO}/pulls/${PR_NUMBER}"
-    declare -a TOKEN_CANDIDATES=()
+    GITEA_API_TOKEN=$(get_gitea_token "$HOST" || true)
    add_token_candidate() {
        local candidate="$1"
        [[ -z "$candidate" ]] && return 0
        local existing
        for existing in "${TOKEN_CANDIDATES[@]:-}"; do
            [[ "$existing" == "$candidate" ]] && return 0
        done
        TOKEN_CANDIDATES+=("$candidate")
    }
-    add_token_candidate "${GITEA_TOKEN:-}"
+    if [[ -n "$GITEA_API_TOKEN" ]]; then
-    add_token_candidate "$(get_gitea_token "$HOST" || true)"
+        RAW=$(curl -sS -H "Authorization: token $GITEA_API_TOKEN" "$API_URL")
-
+    else
-    # Git HTTPS credentials often contain a valid Gitea API token even when a
+        RAW=$(curl -sS "$API_URL")
    # Mosaic credential-source entry is stale. Try them as a fallback before
    # falling back to an unauthenticated request.
    CREDS_FILE="$HOME/.git-credentials"
    if [[ -f "$CREDS_FILE" ]]; then
        while IFS= read -r credential_token; do
            add_token_candidate "$credential_token"
        done < <(grep -F "$HOST" "$CREDS_FILE" 2>/dev/null | sed -n 's#https\?://[^@]*:\([^@/]*\)@.*#\1#p')
    fi
    RAW=""
    HTTP_STATUS=""
    if [[ ${#TOKEN_CANDIDATES[@]} -gt 0 ]]; then
        for GITEA_API_TOKEN in "${TOKEN_CANDIDATES[@]}"; do
            RESPONSE_FILE=$(mktemp)
            HTTP_STATUS=$(curl -sS -o "$RESPONSE_FILE" -w '%{http_code}' -H "Authorization: token $GITEA_API_TOKEN" "$API_URL" || true)
            RAW=$(cat "$RESPONSE_FILE")
            rm -f "$RESPONSE_FILE"
            [[ "$HTTP_STATUS" =~ ^2 ]] && break
        done
    fi
    if [[ ! "$HTTP_STATUS" =~ ^2 ]]; then
        RESPONSE_FILE=$(mktemp)
        HTTP_STATUS=$(curl -sS -o "$RESPONSE_FILE" -w '%{http_code}' "$API_URL" || true)
        RAW=$(cat "$RESPONSE_FILE")
        rm -f "$RESPONSE_FILE"
    fi
    if [[ ! "$HTTP_STATUS" =~ ^2 ]]; then
        ERROR_MESSAGE=$(printf '%s' "$RAW" | python3 -c 'import json, sys
 try:
    data=json.load(sys.stdin)
 except Exception:
    data={}
 print(data.get("message") or data.get("error") or "unknown error")')
        echo "Error: failed to fetch Gitea PR #$PR_NUMBER from $HOST/$OWNER/$REPO (HTTP $HTTP_STATUS): $ERROR_MESSAGE" >&2
        exit 1
    fi
    # Normalize Gitea response to match our expected schema
    METADATA=$(echo "$RAW" | python3 -c "
 import json, sys
 data = json.load(sys.stdin)
 head = data.get('head') or {}
 base = data.get('base') or {}
 user = data.get('user') or {}
 normalized = {
-    'number': data.get('number') or data.get('index'),
+    'number': data.get('number'),
    'title': data.get('title'),
    'body': data.get('body', ''),
    'state': data.get('state'),
-    'author': user.get('login', ''),
+    'author': data.get('user', {}).get('login', ''),
-    'headRefName': head.get('ref') or head.get('label', '').split(':')[-1],
+    'headRefName': data.get('head', {}).get('ref', ''),
-    'baseRefName': base.get('ref') or base.get('label', '').split(':')[-1],
+    'baseRefName': data.get('base', {}).get('ref', ''),
    'labels': [l.get('name', '') for l in data.get('labels', [])],
    'assignees': [a.get('login', '') for a in data.get('assignees', [])],
    'milestone': data.get('milestone', {}).get('title', '') if data.get('milestone') else '',
    'createdAt': data.get('created_at', ''),
    'updatedAt': data.get('updated_at', ''),
-    'url': data.get('html_url') or data.get('url', ''),
+    'url': data.get('html_url', ''),
    'isDraft': data.get('draft', False),
    'mergeable': data.get('mergeable'),
    'diffUrl': data.get('diff_url', ''),
--- a/packages/storage/src/migrate-tier.integration.test.ts
+++ b/packages/storage/src/migrate-tier.integration.test.ts
@@ -16,15 +16,8 @@ import fs from 'node:fs/promises';
 import os from 'node:os';
 import path from 'node:path';
-import {
+import { users, teams, teamMembers, conversations, messages } from '@mosaicstack/db';
-  users,
+import { createPgliteDbWithVector, runPgliteMigrations } from './test-utils/pglite-with-vector.js';
  teams,
  teamMembers,
  conversations,
  messages,
  createPgliteDb,
  runPgliteMigrations,
 } from '@mosaicstack/db';
 import postgres from 'postgres';
 import { afterAll, describe, expect, it } from 'vitest';
@@ -109,8 +102,11 @@ describe.skipIf(!run)('migrate-tier — PGlite → federated PG', () => {
    /* ---- 1. Create a temp PGlite db ---------------------------------- */
    pgliteDataDir = await fs.mkdtemp(path.join(os.tmpdir(), 'fed-m1-08-'));
-    const handle = createPgliteDb(pgliteDataDir);
+    const handle = createPgliteDbWithVector(pgliteDataDir);
-    await runPgliteMigrations(handle);
+
    // Run Drizzle migrations against PGlite.
    // eslint-disable-next-line @typescript-eslint/no-explicit-any
    await runPgliteMigrations(handle.db as any);
    /* ---- 2. Seed representative data --------------------------------- */
--- a/packages/storage/src/test-utils/pglite-with-vector.ts
+++ b/packages/storage/src/test-utils/pglite-with-vector.ts
@@ -0,0 +1,52 @@
 /**
 * Test-only helpers for creating a PGlite database with the pgvector extension
 * and running Drizzle migrations against it.
 *
 * These are intentionally NOT exported from @mosaicstack/db to avoid pulling
 * the WASM vector bundle into the public API surface.
 */
 import { createRequire } from 'node:module';
 import { dirname, resolve } from 'node:path';
 import { PGlite } from '@electric-sql/pglite';
 import { vector } from '@electric-sql/pglite/vector';
 import { drizzle } from 'drizzle-orm/pglite';
 import { migrate as migratePglite } from 'drizzle-orm/pglite/migrator';
 import type { PgliteDatabase } from 'drizzle-orm/pglite';
 import * as schema from '@mosaicstack/db';
 import type { DbHandle } from '@mosaicstack/db';
 /**
 * Create a PGlite DB handle with the pgvector extension loaded.
 * Required for running Drizzle migrations that include `CREATE EXTENSION vector`.
 */
 export function createPgliteDbWithVector(dataDir: string): DbHandle {
  const client = new PGlite(dataDir, { extensions: { vector } });
  const db = drizzle(client, { schema });
  return {
    db: db as unknown as DbHandle['db'],
    close: async () => {
      await client.close();
    },
  };
 }
 /**
 * Run Drizzle migrations against an already-open PGlite database handle.
 * Resolves the migrations folder from @mosaicstack/db's installed location.
 *
 * @param db  A PgliteDatabase instance (from drizzle-orm/pglite).
 */
 export async function runPgliteMigrations(
  // eslint-disable-next-line @typescript-eslint/no-explicit-any
  db: PgliteDatabase<any>,
 ): Promise<void> {
  // Resolve @mosaicstack/db package root to locate its drizzle migrations folder.
  const _require = createRequire(import.meta.url);
  const dbPkgMain = _require.resolve('@mosaicstack/db');
  // dbPkgMain → …/packages/db/dist/index.js  →  dirname = dist/
  // go up one level from dist/ to find the sibling drizzle/ folder
  const migrationsFolder = resolve(dirname(dbPkgMain), '../drizzle');
  await migratePglite(db, { migrationsFolder });
 }
--- a/pnpm-lock.yaml
+++ b/pnpm-lock.yaml
@@ -179,9 +179,6 @@ importers:
      socket.io:
        specifier: ^4.8.0
        version: 4.8.3
      undici:
        specifier: ^7.24.6
        version: 7.24.6
      uuid:
        specifier: ^11.0.0
        version: 11.1.0
@@ -716,10 +713,10 @@ importers:
    dependencies:
      '@mariozechner/pi-agent-core':
        specifier: ^0.63.1
-        version: 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)
+        version: 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)
      '@mariozechner/pi-ai':
        specifier: ^0.63.1
-        version: 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)
+        version: 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)
      '@sinclair/typebox':
        specifier: ^0.34.41
        version: 0.34.48
@@ -6996,6 +6993,10 @@ packages:
    resolution: {integrity: sha512-gBLkYIlEnSp8pFbT64yFgGE6UIB9tAkhukC23PmMDCe5Nd+cRqKxSjw5y54MK2AZMgZfJWMaNE4nYUHgi1XEOw==}
    engines: {node: '>=18.17'}
  undici@7.24.3:
    resolution: {integrity: sha512-eJdUmK/Wrx2d+mnWWmwwLRyA7OQCkLap60sk3dOK4ViZR7DKwwptwuIvFBg2HaiP9ESaEdhtpSymQPvytpmkCA==}
    engines: {node: '>=20.18.1'}
  undici@7.24.6:
    resolution: {integrity: sha512-Xi4agocCbRzt0yYMZGMA6ApD7gvtUFaxm4ZmeacWI4cZxaF6C+8I8QfofC20NAePiB/IcvZmzkJ7XPa471AEtA==}
    engines: {node: '>=20.18.1'}
@@ -7328,6 +7329,12 @@ snapshots:
      '@jridgewell/gen-mapping': 0.3.13
      '@jridgewell/trace-mapping': 0.3.31
  '@anthropic-ai/sdk@0.73.0(zod@3.25.76)':
    dependencies:
      json-schema-to-ts: 3.1.1
    optionalDependencies:
      zod: 3.25.76
  '@anthropic-ai/sdk@0.73.0(zod@4.3.6)':
    dependencies:
      json-schema-to-ts: 3.1.1
@@ -8669,6 +8676,18 @@ snapshots:
      - ws
      - zod
  '@mariozechner/pi-agent-core@0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)':
    dependencies:
      '@mariozechner/pi-ai': 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)
    transitivePeerDependencies:
      - '@modelcontextprotocol/sdk'
      - aws-crt
      - bufferutil
      - supports-color
      - utf-8-validate
      - ws
      - zod
  '@mariozechner/pi-agent-core@0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)':
    dependencies:
      '@mariozechner/pi-ai': 0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)
@@ -8717,6 +8736,30 @@ snapshots:
      - ws
      - zod
  '@mariozechner/pi-ai@0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@3.25.76)':
    dependencies:
      '@anthropic-ai/sdk': 0.73.0(zod@3.25.76)
      '@aws-sdk/client-bedrock-runtime': 3.1008.0
      '@google/genai': 1.45.0(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))
      '@mistralai/mistralai': 1.14.1
      '@sinclair/typebox': 0.34.48
      ajv: 8.18.0
      ajv-formats: 3.0.1(ajv@8.18.0)
      chalk: 5.6.2
      openai: 6.26.0(ws@8.20.0)(zod@3.25.76)
      partial-json: 0.1.7
      proxy-agent: 6.5.0
      undici: 7.24.3
      zod-to-json-schema: 3.25.1(zod@3.25.76)
    transitivePeerDependencies:
      - '@modelcontextprotocol/sdk'
      - aws-crt
      - bufferutil
      - supports-color
      - utf-8-validate
      - ws
      - zod
  '@mariozechner/pi-ai@0.63.2(@modelcontextprotocol/sdk@1.28.0(zod@4.3.6))(ws@8.20.0)(zod@4.3.6)':
    dependencies:
      '@anthropic-ai/sdk': 0.73.0(zod@4.3.6)
@@ -8730,7 +8773,7 @@ snapshots:
      openai: 6.26.0(ws@8.20.0)(zod@4.3.6)
      partial-json: 0.1.7
      proxy-agent: 6.5.0
-      undici: 7.24.6
+      undici: 7.24.3
      zod-to-json-schema: 3.25.1(zod@4.3.6)
    transitivePeerDependencies:
      - '@modelcontextprotocol/sdk'
@@ -12589,7 +12632,7 @@ snapshots:
      saxes: 6.0.0
      symbol-tree: 3.2.4
      tough-cookie: 6.0.1
-      undici: 7.24.6
+      undici: 7.24.3
      w3c-xmlserializer: 5.0.0
      webidl-conversions: 8.0.1
      whatwg-mimetype: 5.0.0
@@ -13309,6 +13352,11 @@ snapshots:
    dependencies:
      mimic-function: 5.0.1
  openai@6.26.0(ws@8.20.0)(zod@3.25.76):
    optionalDependencies:
      ws: 8.20.0
      zod: 3.25.76
  openai@6.26.0(ws@8.20.0)(zod@4.3.6):
    optionalDependencies:
      ws: 8.20.0
@@ -14440,6 +14488,8 @@ snapshots:
  undici@6.21.3: {}
  undici@7.24.3: {}
  undici@7.24.6: {}
  unhomoglyph@1.0.6: {}
--- a/scratchpads/fix-db-bootstrap-migrations.md
+++ b/scratchpads/fix-db-bootstrap-migrations.md
@@ -1,125 +0,0 @@
 # fix(db): bootstrap migrations on local-tier gateway startup
 ## Problem
 Fresh `mosaic gateway install` (npm-installed) leaves the gateway DB schema empty:
 ```
 relation "users" does not exist
 ```
 Sign-in 500s, `auth users create` says "Not signed in", `admin/bootstrap setup`
 also fails — every entry point queries `users` before doing anything else.
 ## Scope
 This PR fixes the **local (PGlite) tier** end-to-end. The postgres-tier path
 has additional pre-existing bugs (see "Known issues, out of scope" below) and
 needs a separate change with real Postgres validation.
 ## Root causes addressed (5 stacked bugs on the local-tier path)
 1. **`packages/db/package.json` `files: ["dist"]`** — the `drizzle/` SQL
   migrations folder is excluded from the published tarball. Even if a
   migrate runner existed, it would have nothing to apply.
 2. **`packages/db/src/migrate.ts`** only supports `postgres-js`. Local-tier
   gateways use embedded PGlite, which can't be reached over a postgres wire
   protocol — so `runMigrations()` is unusable for the local tier.
 3. **`apps/gateway/src/database/database.module.ts`** never invokes
   migrations at startup. The module creates the DB handle and storage
   adapter, but no consumer calls `.migrate()` on either. `mosaic storage
 migrate` CLI even claims "pglite runs schema setup automatically on first
   connection via `adapter.migrate()`" — but `adapter.migrate()` is only
   called by tests, never at runtime.
 4. **`createPgliteDb` does not load the pgvector extension.** Migration 0001
   declares `CREATE EXTENSION IF NOT EXISTS vector;` for the
   `insights.embedding` column. Bare PGlite has no pgvector — the migration
   fails on extension control file lookup.
 5. **Drizzle's PG migrator wraps every migration in one outer transaction.**
   Migration 0009 does `ALTER TYPE grant_status ADD VALUE 'pending'` and then
   `ALTER TABLE federation_grants ALTER COLUMN status SET DEFAULT 'pending'`.
   Postgres' `check_safe_enum_use` rejects the second statement because the
   new enum value isn't committed yet. Splitting the migration into two
   files doesn't help — drizzle batches all migrations into one outer tx.
 ## Fix
 - `packages/db/package.json` — ship `drizzle/` in `files`.
 - `packages/db/src/client-pglite.ts` — load `@electric-sql/pglite/vector`.
 - `packages/db/src/migrate.ts` — add `runPgliteMigrations(handle)`. Walks the
  Drizzle journal and runs each statement-breakpoint chunk through PGlite's
  `client.exec()` (Simple Query protocol → autocommit per statement). Writes
  to the standard `drizzle.__drizzle_migrations` ledger so the result is
  interoperable with `runMigrations()` on a postgres-backed deployment.
  Per-statement try/catch surfaces which statement of which migration failed
  and the ledger row is only written on full success.
 - `packages/db/src/index.ts` — re-export.
 - `apps/gateway/src/database/database.module.ts` — implement `OnModuleInit`:
  - Local tier → `runPgliteMigrations(handle)`, then `storageAdapter.migrate()`
    (the local storage adapter has its own kv tables in a separate PGlite dir).
  - Postgres tier → `storageAdapter.migrate()` only, since
    `PostgresAdapter.migrate()` already calls `runMigrations(url)` against
    the same DATABASE_URL — we deliberately don't double-call.
  NestJS awaits `onModuleInit` before `app.listen()`, so DB-dependent modules
  see a populated schema before any HTTP traffic is accepted.
 - `packages/storage/src/test-utils/pglite-with-vector.ts` — **deleted**.
  The "intentionally not exported" rationale is moot now that migration 0001
  forces pgvector load anyway. `migrate-tier.integration.test.ts` switched
  to `createPgliteDb` + `runPgliteMigrations` from `@mosaicstack/db`.
 ## Tests
 `packages/db/src/migrate.test.ts`:
 - Verifies `runPgliteMigrations` creates the BetterAuth tables (the original
  failure mode).
 - Idempotence (transitively re-runs migration 0009).
 - Partial-failure: pre-creates a conflicting `users` table, asserts the
  thrown error includes statement context (`hash=… statement #N failed`)
  and that no ledger row was written.
 ## QA evidence
 End-to-end on a fresh PGlite install:
 - `[DatabaseModule] Applying PGlite schema migrations...` then
  `Initializing storage adapter (pglite)...` in startup log.
 - `GET /api/bootstrap/status` → `{"needsSetup":true}` HTTP 200 (was 500
  with `relation "users" does not exist`).
 - `POST /api/bootstrap/setup` with empty body → HTTP 400 with Zod
  validation error (was 500), confirming the request reached the
  validator past the table-existence check.
 ## Known issues, out of scope (file separately)
 - **Postgres-tier first install is still broken.** `runMigrations()` uses
  Drizzle's `migratePostgres`, which has the same outer-transaction problem
  as PGlite's migrator. A fresh standalone-tier install would also fail at
  migration 0009. Inline TODO in `migrate.ts:31-35` flags this. Fixing it
  needs either (a) a shared per-statement loop reused for both drivers, or
  (b) splitting migration 0009.
 - **`drizzle/meta/_journal.json` has 0009 ordered before 0008** (`when`
  values `1745280000000` < `1776822435828`). `migratePostgres` skips by
  `created_at < folderMillis`, so on a postgres deployment that already
  applied 0008, 0009 would be skipped forever. Our hash-based skip in the
  PGlite path sidesteps this.
 - **No advisory lock around the migration loop.** Two gateway processes
  pointed at the same DATABASE_URL would race. PGlite is single-process by
  file lock so the local tier is fine; postgres-tier deployments should add
  `pg_advisory_lock(<deterministic-id>)` around the loop in a follow-up.
 - **`mosaic storage migrate` CLI message is misleading** — it claims
  "automatic on first connection via adapter.migrate()" but the adapter
  doesn't self-migrate. With this PR the gateway invokes it explicitly, but
  the CLI message could still be tightened.
 - **Crash mid-migration leaves a partial-state PGlite DB without a ledger
  row.** Detected loudly on next boot (the replay errors on "already
  exists"), but recovery is manual (drop the partially-applied objects or
  insert the migration hash into `drizzle.__drizzle_migrations`). A robust
  fix would add a "started_at" column to a sidecar table to detect
  half-applied state and refuse to start with actionable guidance.