pjandro/OmniRoute
1
0
Fork 0
mirror of https://github.com/diegosouzapw/OmniRoute.git synced 2026-08-19 21:52:21 +03:00
Code Issues Packages Projects Releases Wiki Activity
94 Commits 85 Branches 313 Tags
fa9687ae0db59f49d9549b6621069d9d327d06a8
Commit Graph

26 Commits

Author SHA1 Message Date
diegosouzapw
fa9687ae0d feat: Implement API key usage for EvalsTab LLM calls, enhance model filtering with provider aliases, and remove the random routing strategy option. 2026-02-15 18:56:33 -03:00
diegosouzapw
0ed7738c7c feat: add LLM evaluation framework with golden set testing 
Introduce built-in eval system for testing LLM response quality against
golden sets. Includes eval runner with 4 match strategies (exact,
contains, regex, custom), a pre-loaded 10-case golden set, REST API
endpoints for managing and running suites, and a dashboard UI under
Analytics → Evals with real-time progress tracking and pass rate
summaries. Also adds OpenAPI tags for Audio, Moderations, and Rerank.
 2026-02-15 18:37:18 -03:00
diegosouzapw
a43b1c9218 feat: add random, least-used, and cost-optimized combo strategies 
Extend combo routing with 3 new model selection strategies beyond the
existing priority, weighted, and round-robin options:
- random: Fisher-Yates shuffle for uniform distribution
- least-used: sorts models by request count via comboMetrics
- cost-optimized: sorts models by pricing (cheapest first)

Also enhance the dashboard provider cards with model counts and
clickable detail views for selected providers.
 2026-02-15 16:57:00 -03:00
diegosouzapw
413a9f2a69 feat: add ModelAvailabilityBadge component and retheme landing page 
- Add compact ModelAvailabilityBadge with popover for model status
  monitoring, cooldown clearing, and auto-refresh polling
- Retheme landing page from warm orange tones to cool blue-grey palette
  across FlowAnimation, Footer, and related components
- Update accent color from orange (#f97815) to rose (#E54D5E)
 2026-02-15 16:26:01 -03:00
diegosouzapw
14714aa9f5 feat: refactor dashboard with shared UI component library 
Introduce reusable shared components (Button, Card, Modal, Table,
StatusBadge, EmptyState) and refactor dashboard pages to use them.
Extract client components from server pages, centralize provider
constants, and update CI Node.js matrix from 18 to 20.
 2026-02-15 16:11:56 -03:00
diegosouzapw
06826d4c86 fix(api): resolve TypeError in chat/completions ensureInitialized 
initTranslators() is a no-op stub (translators self-register via static
imports). Wrapping the call with Promise.resolve() ensures .then() works
even when the function returns undefined instead of a Promise.
 2026-02-15 13:24:54 -03:00
diegosouzapw
5c92f05079 fix(ci): resolve build and lint failures 
- Remove non-existent DarkTooltip/CostTooltip exports from analytics/index.js
- Fix setState-in-effect in ProxyTab.js (inline fetch in useEffect)
- Fix anonymous default export warning in prettier.config.mjs
- Remove @ts-check from promptInjectionGuard.js (SanitizeResult type mismatch)
 2026-02-15 12:56:09 -03:00
diegosouzapw
0e238a61fb feat(core): implement 26 action items from critical analysis + bump v0.4.0 
- Security: AES-256-GCM encryption for API keys/tokens, CI security audit
- Accessibility: ARIA labels, aria-live regions, skip-to-content, contrast utility
- Components: Tooltip, CloudSyncStatus, SystemMonitor, StreamTracker
- Utils: costEstimator, promptInjectionGuard middleware, Zod validation schemas
- Docs: openapi.yaml +9 routes, API_REFERENCE internal APIs, version bumps
- Quality: coverage thresholds 60/50/50, error handling improvements
 2026-02-15 12:33:56 -03:00
diegosouzapw
597c590a1d feat(phase10): docs restructuring, component decomposition, and cleanup 
- Split README (1326→211 lines): USER_GUIDE, API_REFERENCE, TROUBLESHOOTING
- Expand CONTRIBUTING.md (112→273 lines)
- Extract RequestLoggerDetail from RequestLoggerV2 (910→665 lines)
- Extract ProxyLogDetail from ProxyLogger (677→519 lines)
- Add accessibility attrs (aria-label, role=dialog) to extracted modals
- Migrate not-found.js inline styles to TailwindCSS gradient classes
- Remove continue-on-error from E2E CI step
- Add PolicyEngine class with glob-based policy matching
- Create a11yAudit.js (WCAG rule checker)
- Fix policyEngine test import path
 2026-02-15 11:53:40 -03:00
diegosouzapw
b08fb31a28 feat(gateway): Phase 9 — LLM Gateway Intelligence 
9.1 — Semantic Cache
  - New: src/lib/semanticCache.js — Two-tier cache (in-memory LRU + SQLite)
  - Signature = SHA-256(model + normalized messages + temperature + top_p)
  - Only caches non-streaming, temperature=0 requests
  - X-OmniRoute-No-Cache: true header bypass
  - Response headers: X-OmniRoute-Cache: HIT/MISS
  - DB table: semantic_cache with indexes on signature and model
  - New: src/app/api/cache/route.js — GET stats, DELETE clear

9.2 — Request Idempotency
  - New: src/lib/idempotencyLayer.js — In-memory 5s dedup window
  - Reads Idempotency-Key or X-Request-Id headers
  - Returns cached response with X-OmniRoute-Idempotent: true header
  - Ephemeral by design (no SQLite)

9.3 — Progress Tracking in Streaming
  - New: open-sse/utils/progressTracker.js
  - Emits SSE 'event: progress' with tokens_generated + elapsed_ms
  - Opt-in via X-OmniRoute-Progress: true header
  - Supports AbortSignal cancellation
  - Final event includes done: true

Integration:
  - chatCore.js: idempotency check → cache check → provider call → cache store → idempotency save
  - Streaming path: optional progress transform chain
  - DB: semantic_cache table added to db/core.js schema

Tests: 320 pass (+25 new) | Build: success
 2026-02-15 11:04:51 -03:00
diegosouzapw
034f1a7e1d feat(pages): Phase 8 — Missing Flows & Pages 
8.1 — 403 Forbidden Page
  - New: src/app/forbidden/page.js
  - Gradient code + Access Denied message + Dashboard link
  - Consistent design with not-found.js

8.2 — Password Recovery Flow
  - New: src/app/forgot-password/page.js
  - Two methods: CLI reset + manual database reset
  - Added 'Forgot password?' link to login page

8.3 — Health/Status Dashboard
  - New: src/app/(dashboard)/dashboard/health/page.js
  - New: src/app/api/monitoring/health/route.js
  - Cards: Uptime, Version, Memory, Provider count
  - Provider health (circuit breaker states with color coding)
  - Rate limit status table
  - Active lockouts list
  - Auto-refresh every 15s
  - Added Health nav link to Sidebar

8.4 — Maintenance Banner
  - New: src/shared/components/MaintenanceBanner.js
  - Auto-detects server health issues every 10s
  - Shows/hides automatically, dismissible
  - Wired into DashboardLayout

8.5 — Empty States
  - Verified EmptyState component applied in 6+ pages
  - Providers page has its own empty handling

Bonus:
  - Fixed stale GitHub link in Sidebar (decolua → diegosouzapw)

Tests: 295 pass | Build: success
 2026-02-15 10:46:11 -03:00
diegosouzapw
7cf8ae8db6 feat(api): Phase 7 — API & Code Quality 
7.1 — Consolidated rate-limit routes
  - Merged rate-limit/ and rate-limits/ into rate-limits/route.js
  - GET returns connections + overview + lockouts + cacheStats (unified)
  - POST handles toggle protection
  - Old rate-limit/ now redirects 308 → rate-limits/
  - Updated frontend fetch URL in providers/[id]/page.js

7.2 — Zod schema for provider constants
  - New: src/shared/validation/providerSchema.js
  - Validates FREE_PROVIDERS, OAUTH_PROVIDERS, APIKEY_PROVIDERS at module load
  - Catches config drift (invalid colors, missing fields) at startup

7.3 — TailwindCSS error pages
  - Converted not-found.js inline styles → Tailwind classes
  - Converted global-error.js inline styles → Tailwind classes
  - Replaced JS hover handlers with Tailwind hover: utilities

7.4 — Fixed GitHub link in privacy page
  - decolua/omniroute → diegosouzapw/OmniRoute

7.5 — Already completed in Phase 5

Tests: 295 pass | Build: success
 2026-02-15 10:15:44 -03:00
diegosouzapw
33c28f73db refactor(phase5-6): domain persistence, policy engine, OAuth extraction, proxy decoupling 
Phase 5 — Foundation & Security:
- SQLite domain state persistence (5 tables, 4 modules: fallback, budget, lockout, circuit breaker)
- Write-through cache pattern for state survival across restarts
- Race condition fix in route.js (Promise-based singleton)
- Default password hardening (.env.example)
- Server init error handling improvement

Phase 6 — Architecture Refactoring:
- OAuth providers extracted into 12 individual modules (providers.js 1051→144 lines)
- Policy Engine (lockout→budget→fallback) with evaluateRequest/evaluateFirstAllowed
- Deterministic round-robin via persistent counter Map
- Telemetry window fix with proper recordedAt timestamps
- Proxy decoupled from API settings (direct import vs HTTP self-fetch)

Tests: 295 pass (22 new: domain-persistence 16, policy-engine 6)
Docs: CHANGELOG, README, ARCHITECTURE.md updated
 2026-02-15 08:12:12 -03:00
diegosouzapw
6e29bd1197 feat(resilience): rate limit overhaul — exponential backoff, circuit breaker, anti-thundering herd, Resilience UI 
Phase 1: Error classification + provider profiles (constants, providerRegistry, accountFallback)
Phase 2: Circuit breaker integration in combo pipeline (combo.js)
Phase 3: Anti-thundering herd mutex + auto rate limits for API key providers (auth.js, rateLimitManager)
Phase 4: Frontend Resilience tab with Circuit Breaker, Provider Profiles, Rate Limit cards

Tests: 63/63 passing (error-classification, combo-circuit-breaker, thundering-herd, rate-limit-enhanced)
 2026-02-15 01:42:50 -03:00
diegosouzapw
5db6676319 feat: detect and handle unrecoverable refresh token errors by marking connections as expired and requiring re-authentication. 2026-02-14 22:04:00 -03:00
diegosouzapw
443e5b7b62 feat(frontend): 100% backend API coverage — 7 batches 
Batch A: Fix Breadcrumbs (usePathname), rewrite ComplianceTab (DataTable+FilterBar+ColumnToggle), delete dead code (a11yAudit.js, policyEngine.js), wire toast notifications (Combos, Providers)
Batch B: ModelAvailabilityPanel — cooldown badges, clear action, auto-refresh
Batch C: BudgetTab — spend cards, progress bars, budget limits form
Batch D: FallbackChainsEditor — color-coded provider chains, create/delete
Batch E: PoliciesPanel — circuit breaker states, locked identifiers, force unlock
Batch F: EvalsTab — expandable suites, run eval, DataTable results
Batch G: TokenHealthBadge + /api/token-health — OAuth health in header

8 new files, 9 modified, 2 deleted. Build passes (exit 0).
 2026-02-14 21:04:51 -03:00
diegosouzapw
e3fc9387be feat(ui): wire budget/telemetry/compliance into dashboard pages 
Batch 4 — Page Integration:
- Usage page: add BudgetTelemetryCards (latency p50/p95/p99, cache, system health)
- Settings page: add ComplianceTab (audit log), CacheStatsCard (prompt cache + flush)
- Combos page: replace inline empty state with EmptyState component

3 new components, 3 page modifications. Build verified: exit code 0
 2026-02-14 20:21:15 -03:00
diegosouzapw
7e066df6cd feat(api): create 9 API routes for backend module access 
Batch 2 — API Routes:
- /api/cache/stats — GET cache stats, DELETE flush
- /api/models/availability — GET availability report, POST clear cooldown
- /api/telemetry/summary — GET p50/p95/p99 latency metrics
- /api/usage/budget — GET cost summary, POST set budget per key
- /api/fallback/chains — GET/POST/DELETE fallback chain management
- /api/compliance/audit-log — GET filterable audit log
- /api/evals — GET list suites, POST run suite
- /api/evals/[suiteId] — GET suite details
- /api/policies — GET circuit breaker + lockout status, POST force-unlock

Build verified: exit code 0
 2026-02-14 20:11:27 -03:00
diegosouzapw
f77dd89d53 feat(remaining): deferred tasks — error pages, UX components, telemetry, domain extraction 
T-20 — .gitignore cleanup:
- Add .analysis/ and antigravity-manager-analysis/ to gitignore
- Whitelist FASE docs, PLANO-IMPLANTACAO.md, TASKS.md

T-21 — Error pages:
- Create not-found.js (404 page with gradient design)
- Create global-error.js (root error boundary with dev details)

T-36 — Breadcrumbs:
- Create Breadcrumbs.js with path-to-label mapping and ARIA semantics

T-37 — Empty states:
- Create EmptyState.js with bounce animation and optional CTA

T-45 — Request telemetry:
- Create requestTelemetry.js (7-phase lifecycle, p50/p95/p99 aggregation)

T-46 — Domain extraction:
- Create comboResolver.js (priority/round-robin/random/least-used strategies)
- Create lockoutPolicy.js (sliding window lockout with force-unlock)

Tests: 13/13 new tests pass (88/88 total)
 2026-02-14 18:43:07 -03:00
diegosouzapw
81a4f2986c feat: v0.2.0 — advanced routing services, cost analytics dashboard, pricing overhaul 
Added:
- 8 new open-sse services (account selector, IP filter, session manager, etc.)
- 6 new dashboard settings tabs (IP filter, system prompt, thinking budget, pricing)
- Usage cost dashboard with provider cost donut, cost trend line, model cost column
- Pricing API merging registry + custom + pricing-only models
- 9 unit tests for all new services

Changed:
- Usage analytics layout redesigned with prominent cost display
- DailyTrendChart upgraded to ComposedChart with dual Y-axes

Fixed:
- Pricing page now shows custom/imported models
- Icon rendering (material-symbols-rounded → outlined)
 2026-02-14 13:50:45 -03:00
diegosouzapw
c6000ecc8e feat(providers): add new endpoints (rerank, audio, moderations) and providers (Hyperbolic, Deepgram, AssemblyAI, NanoBanana) 
- Add /v1/rerank endpoint with Cohere, Together, NVIDIA, Fireworks
- Add /v1/audio/transcriptions with OpenAI, Groq, Deepgram, AssemblyAI
- Add /v1/audio/speech with OpenAI, Hyperbolic, Deepgram
- Add /v1/moderations with OpenAI
- Add Hyperbolic as chat provider (8 models, OpenAI-compatible)
- Add Hyperbolic image generation (SDXL, SD2, FLUX)
- Add NanoBanana image generation (Flash + Pro via nanobananaapi.ai)
- Add Deepgram STT (Nova 3, Nova 2) with Token auth and binary format
- Add AssemblyAI STT (Universal 3 Pro) with async upload-poll workflow
- Add Deepgram TTS (Aura voices) and Hyperbolic TTS (Melo)
- Update EndpointPageClient to show 7 endpoint sections
- Update /v1/models to return type/subtype for all model categories
- Fix build: remove output:standalone from next.config.mjs
 2026-02-13 22:57:32 -03:00
diegosouzapw
c0c2816671 feat: Implement model selection and dynamic model fetching in ChatTesterMode and TestBenchMode. 2026-02-13 21:11:19 -03:00
diegosouzapw
2b1b8e4539 chore(ui): rebrand to OmniRoute 
- Sidebar: 'Endpoint Proxy' → 'OmniRoute'
- Page title: 'OmniRoute — AI Gateway for Multi-Provider LLMs'
- Description updated across layout and config
 2026-02-13 19:13:57 -03:00
diegosouzapw
0af22a558f fix(oauth): prevent connection test from corrupting valid tokens 
Only attempt token refresh on 401/403 during connection tests when
the token is actually expired (isTokenExpired). Previously, any 401/403
triggered an aggressive refresh that could overwrite valid tokens when
the upstream returned transient errors (rate-limiting, etc.).

Fixes Cline, Qwen, and iFlow losing authentication after tests.
 2026-02-13 18:09:28 -03:00
diegosouzapw
50fc931086 feat(versioning): set initial version to 0.0.1 and restore dynamic model import 
- Set package.json version to 0.0.1 (initial OmniRoute release)
- Set open-sse/package.json version to 0.0.1
- Restore 'Import from /models' button for standard providers (openai, gemini, deepseek, etc.)
- Add handleImportModels function to ProviderDetailPage
 2026-02-13 16:49:06 -03:00
diegosouzapw
699329944e feat: initial OmniRoute release (rebranded from 9router) 
This project is inspired by and originally forked from 9router by decolua
(https://github.com/decolua/9router).

Full rebrand: 9router → OmniRoute across all source code, configuration,
Docker, documentation, and assets.
 2026-02-13 16:29:27 -03:00