20 Commits

Author	SHA1	Message	Date
Max Mayfield	5792f95d7c	Fix BMad adversarial security review findings ... Resolves 11 of the 13 findings: - [CRITICAL] SQLi in RLS: replaced SET LOCAL with parameterized set_config() - [CRITICAL] Rate Limiting: installed and registered @fastify/rate-limit in all 5 apps - [CRITICAL] Invite Hijacking: added email verification check to invite lookup - [HIGH] Webhook HMAC: added Fastify rawBody parser to fix JSON.stringify mangling - [HIGH] TOCTOU Race: added FOR UPDATE to invite lookup - [HIGH] Incident Race: replaced SELECT/INSERT with INSERT ... ON CONFLICT - [MEDIUM] Grafana Timing Attack: replaced === with crypto.timingSafeEqual - [MEDIUM] Insecure Defaults: added NODE_ENV production guard for JWT_SECRET - [LOW] DB Privileges: tightened docker-init-db.sh grants (removed ALL PRIVILEGES) - [LOW] Plaintext Invites: tokens are now hashed (SHA-256) before DB storage/lookup - [LOW] Scrypt: increased N parameter to 65536 for stronger password hashing Note: - Finding #4 (Fragmented Identity) requires a unified auth database architecture. - Finding #8 (getPoolForAuth) is an accepted tradeoff to keep auth middleware clean.	2026-03-03 00:14:39 +00:00
Max Mayfield	eb953cdea5	Security hardening: auth encapsulation, pool restriction, rate limiting, invites, async webhooks ... Phase 1 (Security Critical): - Auth plugin encapsulation: replaced global addHook with Fastify plugin scope - Removed startsWith URL matching; public routes registered outside auth scope - JWT verify now enforces algorithms: ['HS256'] (prevents algorithm confusion) - Raw pool no longer exported from db.ts; systemQuery() + getPoolForAuth() instead - withTenant() remains primary tenant-scoped query path Phase 2 (Infrastructure): - docker-compose.yml: all secrets via env var substitution (${VAR:-default}) - Per-service Postgres users (dd0c_drift, dd0c_alert, etc.) in docker-init-db.sh - .env.example with all configurable secrets - build-push.sh uses $REGISTRY_PASSWORD instead of hardcoded - .gitignore excludes .env files - @fastify/rate-limit: 100 req/min global, 5/min login, 3/min signup - CORS_ORIGIN default changed from '*' to 'http://localhost:5173' Phase 3 (Product): - Team invite flow: tenant_invites table, POST /invite, GET /invites, DELETE /invites/:id - Signup accepts optional invite_token to join existing tenant - Async webhook ingestion (P3): LPUSH to Redis, BRPOP worker, dead-letter queue Console: - All 5 product modules wired: drift, alert, portal, cost, run - PageHeader accepts children prop - 71 modules, 70KB gzipped production build All 6 projects compile clean (tsc --noEmit).	2026-03-02 23:53:55 +00:00
Max Mayfield	be3f37cfdd	Fix CRITICAL auth bypass: exact match for login/signup paths ... startsWith('/api/v1/auth/login') allowed any path with that prefix to bypass authentication (e.g. /api/v1/auth/login-anything). Changed to exact path match with query string stripping. Fixed across all 5 products + shared/auth.ts.	2026-03-02 20:35:28 +00:00
Max Mayfield	3be37d1293	Skip auth on /version endpoint (same as /health)	2026-03-02 13:54:46 +00:00
Max Mayfield	5bad2481ae	Add /version endpoint to all products + BUILD_SHA/BUILD_TIME in Dockerfiles	2026-03-02 13:53:15 +00:00
Max Mayfield	2c9408b1df	Fix indentation on registerWebhookSecretRoutes call	2026-03-02 05:11:37 +00:00
Max Mayfield	9f46b84257	Add API to manage webhook secrets for alert integrations	2026-03-02 05:00:40 +00:00
Max Mayfield	81d03c1735	Fix tenant slug collision: append random hex suffix to prevent 23505 on duplicate tenant names	2026-03-01 22:36:21 +00:00
Max Mayfield	27a89ee2b7	Trigger CI with tsc fix	2026-03-01 06:56:00 +00:00
Max Mayfield	bfc599da52	Trigger CI after workflow rewrite	2026-03-01 06:47:59 +00:00
Max Mayfield	68140881e0	Trigger CI for P3-P6 Node products	2026-03-01 06:43:58 +00:00
Max Mayfield	4534f0aeba	Fix test failures: HMAC length check (P3), fast-check fround (P5) ... - P3: timingSafeEqual requires equal-length buffers; add length guard before compare - P5: fast-check fc.float requires 32-bit floats; wrap min with Math.fround() - All 5 Node products: 83 tests passing across 13 test files	2026-03-01 06:24:46 +00:00
Max Mayfield	4146f1c4d0	Fix TypeScript compilation errors across P3-P6 ... - jwt.sign: explicit SignOptions cast for expiresIn (all 4 products) - ioredis: use named import { Redis } instead of default (P4, P6) - P4 catalog/service: fix import paths for aws-scanner and github-scanner - P4 discovery: pass pool to ScheduledDiscovery constructor - P6 agent-bridge: add explicit types for Redis message callback params - All 4 Node products now compile cleanly with tsc --noEmit	2026-03-01 06:06:31 +00:00
Max Mayfield	e1b22e5309	Wire up remaining TODO stubs: P3 test notifications, P2 drift notification trigger ... - P3: test notification endpoint now instantiates real Slack/Email/Webhook notifiers - P2: drift processor triggers notification service when drift_score > 0 (non-fatal on failure)	2026-03-01 04:14:26 +00:00
Max Mayfield	5ee869b9d8	Implement auth: login/signup (scrypt), API key generation, shared migration ... - Login: email + password lookup, scrypt verify, JWT token - Signup: create tenant + owner user in transaction, slug generation - API key: dd0c_ prefix, SHA-256 hash (not bcrypt — faster for API key lookups), prefix index - Scrypt over bcrypt: zero native deps, Node.js built-in crypto - Auth routes skip JWT middleware (login/signup are public) - 002_auth.sql: users + api_keys tables with RLS, copied to all products - Synced auth middleware to P3/P4/P5/P6	2026-03-01 03:19:18 +00:00
Max Mayfield	bdaa732ce1	Implement TODO stubs: webhook secret lookup, alert→incident wiring, catalog upsert/stage ... - P3: getWebhookSecret() now queries DB; ingestAlert() creates/attaches incidents, auto-resolves on resolved status - P4: stageUpdates() writes to staged_updates table; upsertService() with ON CONFLICT; getService/updateOwner implemented	2026-03-01 03:18:05 +00:00
Max Mayfield	829e408e1e	Add notification dispatchers (P3 Slack/Email/Webhook, P5 Slack), full YAML parser for P6 ... - P3 alert: NotificationDispatcher with Slack Block Kit, Resend email, generic webhook; severity-gated dispatch - P5 cost: CostSlackNotifier with anomaly Block Kit (score, deviation, snooze/expected buttons) - P6 run: Full YAML runbook parser with serde_yaml, variable substitution ({{var}}), failure actions, 7 tests - P6 parser: validates non-empty steps, default timeout (300s), default abort on failure	2026-03-01 03:13:06 +00:00
Max Mayfield	f2e0a32cc7	Wire auth middleware into all products, add docker-compose and init-db script ... - Auth middleware (JWT + API key + RBAC) copied into P3/P4/P5/P6 - All server entry points now register auth hooks + auth routes - Webhook and Slack endpoints skip JWT auth (use HMAC/signature) - docker-compose.yml: shared Postgres + Redis + Meilisearch, all 4 Node products as services - init-db.sh: creates per-product databases and runs migrations - P1 (Rust) and P2 (Go agent) run standalone, not in compose	2026-03-01 03:10:35 +00:00
Max Mayfield	d85cdaa3e7	Flesh out dd0c/alert: webhook routes, incident API, notification config, data layer ... - Webhook routes: Datadog, PagerDuty, OpsGenie, Grafana with per-tenant HMAC/token auth - Incident API: list (filtered), detail with alerts, acknowledge/resolve/suppress, dashboard summary - Notification config: CRUD with upsert, test endpoint, Slack/email/webhook channels - Grafana normalizer: severity mapping (critical/warning/info) - Data layer: withTenant() RLS wrapper, Zod config validation - Fastify server entry point with cors/helmet	2026-03-01 03:04:57 +00:00
Max Mayfield	ccc4cd1c32	Scaffold dd0c/alert: ingestion, correlation engine, HMAC validation, tests ... - Webhook ingestion: HMAC validation for Datadog/PagerDuty/OpsGenie with 5-min timestamp freshness - Payload normalizers: canonical alert schema with severity mapping per provider - Correlation engine: time-window grouping, late-alert attachment (2x window), FakeClock for testing - InMemoryWindowStore for unit tests - Tests: 12 HMAC validation cases, 5 normalizer cases, 7 correlation engine cases - PostgreSQL schema with RLS: tenants, incidents, alerts, webhook_secrets, notification_configs - Free tier enforcement columns (alert_count_month, reset_at) - Fly.io config, Dockerfile, Gitea Actions CI	2026-03-01 02:49:14 +00:00