pjandro/OmniRoute
1
0
Fork 0
mirror of https://github.com/diegosouzapw/OmniRoute.git synced 2026-07-31 20:32:20 +03:00
Code Issues Packages Projects Releases Wiki Activity
746 Commits 94 Branches 313 Tags
v2.7.3
Commit Graph

746 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
diegosouzapw
896ce9c0e2 feat(release): v2.7.3 — fix Codex direct API weekly quota fallback 
- fix(codex): resolveQuotaWindow() prefix-matches 'weekly' → 'weekly (7d)' cache keys
- fix(codex): applyCodexWindowPolicy() enforces useWeekly/use5h toggles in direct API
- 4 new regression tests, 766 total passing
- Closes #440
v2.7.3 		2026-03-18 08:41:13 -03:00
Diego Rodrigues de Sa e Souza
82934132e9 Merge pull request #441 from rexname/fix/issue-440-direct-api-fallback 
fix(codex): block weekly-exhausted accounts in direct API fallback
 2026-03-18 08:40:19 -03:00
rexname
a2012b70de chore(review): harden window normalization and deterministic quota matching 2026-03-18 14:17:37 +07:00
rexname
bcfeba8a57 fix(codex): enforce weekly quota blocking for direct API fallback 2026-03-18 13:57:25 +07:00
diegosouzapw
d3dfd9ce57 feat(release): v2.7.2 — fix light mode contrast in logs UI 
- fix(logs): text colors in filter buttons + combo badge now have dark: variants
- Bumped version to 2.7.2
- Updated CHANGELOG and openapi.yaml
v2.7.2 		2026-03-18 00:42:22 -03:00
Diego Rodrigues de Sa e Souza
aa06d5d356 Merge pull request #433 from diegosouzapw/fix/issue-378-logs-light-mode-contrast 
Merged fix for light mode contrast in filter buttons and combo badge. Thanks @rdself for the great bug report!
 2026-03-18 00:41:28 -03:00
diegosouzapw
448c8a29e1 fix(logs): fix light mode contrast in filter buttons and combo badge (#378) 
- text-red-400 → text-red-700 dark:text-red-400 (error filter, recording button)
- text-emerald-400 → text-emerald-700 dark:text-emerald-400 (ok filter)
- text-violet-300 → text-violet-700 dark:text-violet-300 (combo filter)
- combo row badge: violet-700 → violet-800 dark:violet-300, stronger border

Fixes #378
 2026-03-17 16:46:27 -03:00
diegosouzapw
928b7120f4 feat(release): v2.7.1 — unified web search routing + Next.js 16.1.7 security 
- POST /v1/search: 5 providers (Serper, Brave, Perplexity, Exa, Tavily), 6,500+ free/mo
- Search analytics dashboard tab + GET /api/v1/search/analytics
- db: request_type column on call_logs (migration 007)
- Next.js 16.1.7: 6 CVEs fixed (critical: CVE-2026-29057 HTTP request smuggling)
- docs/openapi.yaml: bumped to 2.7.1
v2.7.1 		2026-03-17 16:27:31 -03:00
diegosouzapw
a3deacd718 feat: Implement historical model latency and success rate tracking for auto-combo routing and update Claude and Deepseek pricing and model registrations. 2026-03-17 16:18:36 -03:00
diegosouzapw
78959fffbd Merge branch 'main' of https://github.com/diegosouzapw/OmniRoute 2026-03-17 16:18:12 -03:00
Diego Rodrigues de Sa e Souza
1788616e52 Merge pull request #431 from diegosouzapw/dependabot/npm_and_yarn/next-16.1.7 
Security update merged: Next.js 16.1.7 fixes 6 CVEs including critical CVE-2026-29057 (HTTP request smuggling). No breaking changes.
 2026-03-17 16:18:01 -03:00
Diego Rodrigues de Sa e Souza
c61e6d0777 Merge pull request #432 from Regis-RCR/feat/search-provider-routing 
Merged with dashboard improvements: SearchAnalyticsTab + /api/v1/search/analytics endpoint — PR review complete by Antigravity.
 2026-03-17 16:17:39 -03:00
diegosouzapw
a3bc7620b1 feat(integration): integrate ClawRouter services into active pipeline 
- intentClassifier → engine.ts selectProvider()
  When taskType is 'default', classifies prompt via multilingual keyword
  detection (9 langs) and uses detected intent (code/reasoning/simple/medium)
  for 6-factor task fitness scoring.

- emergencyFallback → chatCore.ts error path (after T5 intra-family fallback)
  On HTTP 402 or budget-exhaustion keywords, attempts one redirect to
  nvidia/gpt-oss-120b ($0.00/M) before returning error to combo router.
  Skipped for streaming requests and tool-calling requests.

- AutoComboConfig.routerStrategy field added
  Allows per-combo strategy override ('rules' | 'cost' | 'latency')

Note: requestDedup was already integrated in chatCore.ts (line 387-430)
Branch: feat/clawrouter-improvements
 2026-03-17 15:22:12 -03:00
diegosouzapw
8064c588dc docs(i18n): sync v2.7.0 release notes to 29 language READMEs 
New in v2.7.0: pluggable RouterStrategy, multilingual intent detection,
request deduplication, new providers (Grok-4 Fast, GLM-5/Z.AI,
MiniMax M2.5, Kimi K2.5). Native translations for de/es/fr/it/ru/zh-CN/ja/ko/ar/pt-BR/pt.
 2026-03-17 15:11:09 -03:00
Regis
564e983c68 feat(search): add unified web search routing with 5 providers 
Add POST /v1/search — a unified search endpoint routing queries across
5 providers (Serper, Brave, Perplexity Search, Exa, Tavily) with
automatic failover, in-memory caching, and request coalescing.

No open-source AI gateway offers unified search routing. This chains
free tiers for 5,500+ searches/month with zero downtime.

Providers: Serper ($0.001/q, 2500/mo free), Brave ($0.005/q, 1000/mo),
Perplexity Search ($0.005/q), Exa ($0.007/q, 1000/mo), Tavily
($0.008/q, 1000/mo). Auto-select picks cheapest with credentials.

Architecture follows existing patterns:
- searchRegistry.ts (same as embeddingRegistry.ts)
- search.ts handler (same as embeddings.ts)
- route.ts (same as /v1/embeddings/route.ts)
- searchCache.ts (bounded TTL cache + request coalescing)

Schema finalized — all future fields defined as optional with safe
defaults. No breaking changes when implementing content extraction,
answer synthesis, or ranking.

Key features:
- Per-provider request builders and response normalizers
- Enriched response: display_url, score, favicon_url, content block,
  metadata, answer block, errors array, upstream_latency_ms metrics
- Cost-sorted auto-select with failover on 429/5xx/timeout
- Credential fallback (perplexity-search reuses perplexity chat key)
- Cache key includes all result-affecting parameters
- max_results clamped to provider limits, sanitized error responses
- Factored validators (validateSearchProvider factory)
- CORS headers on all responses
- Dashboard: Search & Discovery section, search provider template
- DB migration 007: request_type column in call_logs
- 28 unit tests (registry, cache, coalescing, validation)
 2026-03-17 18:28:35 +01:00
diegosouzapw
e1da181740 fix(publish): also remove app/electron/ (contains app.asar binary) to prevent Z_DATA_ERROR 2026-03-17 14:25:48 -03:00
diegosouzapw
c63209200e fix(publish): remove app/vscode-extension/ after build to prevent Z_DATA_ERROR in npm pack 2026-03-17 14:13:15 -03:00
diegosouzapw
737808cf53 fix(npm): exclude app/vscode-extension/ from package to prevent Z_DATA_ERROR during publish 2026-03-17 13:50:06 -03:00
diegosouzapw
a197bb7736 fix(routerStrategy): use .ts extension in imports for Next.js App Router bundle compatibility 2026-03-17 13:15:47 -03:00
dependabot[bot]
f9dd967bc5 deps: bump next from 16.1.6 to 16.1.7 
Bumps [next](https://github.com/vercel/next.js) from 16.1.6 to 16.1.7.
- [Release notes](https://github.com/vercel/next.js/releases)
- [Changelog](https://github.com/vercel/next.js/blob/canary/release.js)
- [Commits](https://github.com/vercel/next.js/compare/v16.1.6...v16.1.7)

---
updated-dependencies:
- dependency-name: next
  dependency-version: 16.1.7
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
 2026-03-17 16:14:44 +00:00
diegosouzapw
44e4d55a66 feat(release): merge feat/clawrouter-improvements — v2.7.0 v2.7.0 2026-03-17 13:12:41 -03:00
diegosouzapw
095c84ac16 fix(providerRegistry): remove duplicate claude-haiku-4-5-20251001 from anthropic provider to prevent ambiguous model resolution 2026-03-17 13:10:23 -03:00
diegosouzapw
e063eae727 feat(clawrouter): implement 14 ClawRouter-inspired features 
PRICING UPDATES (01-09):
- xAI Grok-4 family: grok-4-fast-non-reasoning (/usr/bin/bash.20/$0.50/M, 1143ms),
  grok-4-fast-reasoning, grok-4-1-fast-*, grok-4-0709, grok-3, grok-3-mini
- Z.AI GLM-5 family: glm-5 + glm-5-turbo (128k maxOutput, $1.00/$3.20/M)
- Gemini Flash Lite: price corrected $0.15→$0.10 / $1.25→$0.40 (per ClawRouter)
- Gemini 3.1 Pro: new flagship (1.05M context, aliased as gemini-3.1-pro)
- Anthropic Claude 4.5/4.6: haiku-4.5 ($1/$5), sonnet-4.6 ($3/$15), opus-4.6 ($5/$25)
- DeepSeek native section: deepseek-chat/v3/v3.2 ($0.28/$0.42), deepseek-reasoner ($0.55/$2.19)
- Kimi K2.5 Moonshot: kimi-k2.5 ($0.60/$3.00, 262k ctx), moonshot-kimi-k2.5 alias
- MiniMax M2.5: minimax-m2.5 ($0.30/$1.20, 204k ctx, reasoning+tools)
- NVIDIA free tier: gpt-oss-120b at $0.00/M via emergencyFallback.ts

INFRASTRUCTURE FEATURES (10-14):
- feat(router): add intentClassifier.ts for multilingual intent detection (9 langs)
  Detects code/reasoning/simple in EN, PT-BR, ES, ZH, JA, RU, DE, KO, AR
- feat(dedup): add requestDedup.ts for concurrent request deduplication
  SHA-256 hash, skip streaming, skip high-temperature, 60s failsafe TTL
- feat(autoCombo): add routerStrategy.ts pluggable strategy system
  RouterStrategy interface, RulesStrategy (6-factor) + CostStrategy, registry
- feat(fallback): add emergencyFallback.ts budget-exhaustion detector
  Triggers on HTTP 402 or budget keywords, redirects to nvidia/gpt-oss-120b
- feat(taskFitness): add fitness scores for Grok-4, Kimi K2.5, GLM-5,
  MiniMax M2.5, DeepSeek V3.2, Gemini 3.1 Pro across all task categories

PROVIDERS:
- providers.ts: add Z.AI (zai) provider entry for GLM-5 API key connections

All features on branch: feat/clawrouter-improvements
Source: github.com/BlockRunAI/ClawRouter analysis (2026-03-17)
 2026-03-17 10:43:12 -03:00
diegosouzapw
f02c5b5c69 fix(install/v2.6.10): Windows better-sqlite3 prebuilt download (#426) 
npm version patch run BEFORE staging files — this is an ATOMIC commit.

Adds Strategy 1.5 to scripts/postinstall.mjs:
- Uses @mapbox/node-pre-gyp install --fallback-to-build=false
  (bundled within better-sqlite3) to download the correct prebuilt
  binary for the current OS/arch (win32-x64/arm64, darwin-x64/arm64)
  WITHOUT requiring node-gyp, Python, or MSVC build tools.
- Tries node-pre-gyp.cmd (Windows) or node-pre-gyp (Unix) from .bin/
  with fallback to direct path in @mapbox/node-pre-gyp/bin/
- Falls back to npm rebuild only if prebuilt download fails.
- Windows-specific error: shows Option A (npx node-pre-gyp) and
  Option B (rebuild) with Visual Studio Build Tools links.

Fixes: #426 (better_sqlite3.node is not a valid Win32 application)
v2.6.10 		2026-03-17 10:09:45 -03:00
diegosouzapw
838f1d645c fix(v2.6.9): CI budget checks, #409 file attachments, atomic release workflow 
Includes version bump — v2.6.9 — committed ATOMICALLY with all changes:

fixes:
- fix(ci/t11): Remove 'any' from comments in openai-responses.ts + chatCore.ts
  (\bany\b regex counted comment text as explicit any violations)
- fix(chatCore/#409): Normalize unsupported content part types before forwarding
  Cursor sends {type:'file'} for .md attachments; Copilot/OpenAI providers reject
  with 'type has to be either image_url or text'. Now: file/document→text block,
  unknown types dropped with debug log. Fixes claude-* models via github-copilot.

workflow:
- chore(generate-release): ATOMIC COMMIT RULE — npm version patch MUST run before
  feature commits so the release tag always points to a commit with full changes
v2.6.9 		2026-03-17 09:09:01 -03:00
diegosouzapw
ce2c30c437 chore(release): v2.6.8 — combo agents, auto-update, detailed logs, MITM Kiro v2.6.8 2026-03-17 08:58:03 -03:00
diegosouzapw
d56fae0a7b feat: combo agents, auto-update UI, detailed logs, MITM Kiro (#399 #401 #320 #378 #336) 
DB Migrations (zero-breaking, ADD COLUMN DEFAULT NULL + new table):
- 005_combo_agent_fields.sql: system_message, tool_filter_regex, context_cache_protection on combos
- 006_detailed_request_logs.sql: ring-buffer table (500 entries) for full pipeline body capture

Features:
- #399 System Message Override + Tool Filter Regex per Combo
  - applyComboAgentMiddleware() injected into handleComboChat/handleRoundRobinCombo
  - Supports both OpenAI and Anthropic tool name formats
- #401 Context Caching Protection (Stateless)
  - injectModelTag() appends <omniModel>provider/model</omniModel> to responses
  - extractPinnedModel() reads tag from history and pins model for session
- #320 Auto-Update via Settings
  - GET /api/system/version — current vs latest npm
  - POST /api/system/update — fire-and-forget npm install + pm2 restart
- #378 Detailed Request Logs
  - saveRequestDetailLog() captures bodies at 4 pipeline stages (opt-in toggle)
  - GET/POST /api/logs/detail — list logs + enable/disable toggle
- #336 MITM Kiro IDE
  - src/mitm/targets/kiro.ts: MitmTarget profile for api.anthropic.com interception
 2026-03-17 08:53:41 -03:00
diegosouzapw
e45ef00bef chore(release): v2.6.7 — SSE fixes, local provider_nodes, proxy registry 
PRs merged: #414 (deps) #415 #417 #419 #420 #421 (SSE fixes)
            #418 (Claude passthrough) #422 #416 #423 (local nodes)
            #427 (strip empty blocks) #428 (OAuth refreshable)
            #429 (proxy registry)
Contributors: @prakersh, @Regis-RCR, @dependabot
v2.6.7 		2026-03-17 08:17:11 -03:00
diegosouzapw
e9f31f7394 Merge pull request #429 from contributor branch 2026-03-17 08:14:05 -03:00
diegosouzapw
7c10a98eb2 Merge pull request #428 from contributor branch 2026-03-17 08:14:04 -03:00
diegosouzapw
f260483101 Merge pull request #427 from contributor branch 2026-03-17 08:14:03 -03:00
diegosouzapw
389e6e5c9e Merge pull request #423 from contributor branch 2026-03-17 08:14:02 -03:00
diegosouzapw
1cfd5866be Merge pull request #422 from contributor branch 2026-03-17 08:14:02 -03:00
diegosouzapw
c7ceac7f41 Merge pull request #421 from contributor branch 2026-03-17 08:14:01 -03:00
diegosouzapw
cd6eca0424 Merge pull request #420 from contributor branch 2026-03-17 08:14:00 -03:00
diegosouzapw
8c6136fea0 fix(sse): generate fallback call_id for tool calls with missing IDs (#419) 
Co-authored-by: Prakersh Maheshwari <prakersh@users.noreply.github.com>
 2026-03-17 08:11:53 -03:00
Diego Rodrigues de Sa e Souza
9644444028 Merge pull request #418 from prakersh/fix/claude-to-claude-passthrough 
fix(sse): add Claude-to-Claude passthrough for anthropic-compatible providers
 2026-03-17 08:09:44 -03:00
Diego Rodrigues de Sa e Souza
9c4154291d Merge pull request #417 from prakersh/fix/orphaned-tool-result-filter 
fix(sse): filter orphaned tool results after context compaction
 2026-03-17 08:09:41 -03:00
Diego Rodrigues de Sa e Souza
533f5f6da6 Merge pull request #416 from Regis-RCR/feat/audio-provider-nodes 
feat(audio): route audio requests to local provider_nodes
 2026-03-17 08:09:38 -03:00
Diego Rodrigues de Sa e Souza
1b8de756cd Merge pull request #415 from prakersh/fix/empty-tool-name-loop 
fix(sse): skip empty-name tool calls in Responses API translator
 2026-03-17 08:09:28 -03:00
Diego Rodrigues de Sa e Souza
650b415537 Merge pull request #414 from diegosouzapw/dependabot/npm_and_yarn/development-cc00f57801 
deps: bump the development group with 4 updates
 2026-03-17 08:09:25 -03:00
rexname
04b50329fc fix(proxy): address PR review findings for auth, credentials, and health stats 2026-03-17 16:58:44 +07:00
Regis
25aab8c55c feat(audio): route audio requests to local provider_nodes 
Audio endpoints (/v1/audio/speech and /v1/audio/transcriptions) only
supported hardcoded providers from audioRegistry.ts. Local inference
backends configured as provider_nodes (e.g., MLX-Audio, oMLX) could
not serve audio through OmniRoute.

This adds a Phase 3 fallback in the audio model parser that consults
provider_nodes from the database. Local providers with api_type=openai
are automatically available for audio routing via their prefix
(e.g., mlx-audio/tts-model, omlx/whisper-large-v3-turbo).

Design: injection pattern — Next.js route handlers load provider_nodes
(async DB query) and pass them to the sync parser as a parameter.
No cross-workspace imports, no breaking changes to existing parsers.

Changes:
- Add buildDynamicAudioProvider() in audioRegistry.ts
- Add Phase 3 (provider_nodes prefix match) to parseAudioModel()
- Extend parseSpeechModel/parseTranscriptionModel with optional
  dynamicProviders parameter (backward compatible)
- Load and inject provider_nodes in speech/transcription route handlers
- Dynamic providers use authType=none (local, no credentials needed)
 2026-03-17 09:24:18 +01:00
Oleg Saprykin
ceda2e70c1 fix(api): add refreshable: true to claude OAuth test config 
Claude OAuth tokens are short-lived and require refresh. The runtime
HealthCheck (open-sse) already refreshes them successfully, but the
Dashboard test endpoint was missing `refreshable: true` in its config.

This caused the Dashboard to show "auth failed / Token expired" for
Claude providers even though the tokens were being refreshed correctly
at runtime. The codex provider already had this flag set.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-17 10:47:35 +03:00
Oleg Saprykin
2908303d4b fix(sse): strip empty text content blocks before translation 
Anthropic API rejects requests containing {"type":"text","text":""} with
400 "text content blocks must be non-empty". Some clients like LiteLLM
passthrough and @ai-sdk/anthropic may forward empty text blocks as-is.

Filter out empty text content blocks from messages before calling
translateRequest, similar to how empty-name tools are already stripped.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-17 10:46:24 +03:00
diegosouzapw
a9f69711c6 fix(build): remove node: protocol prefix from all src/ imports (#turbopack-compat) 
Turbopack (Next.js 15) does not process node: URL prefixes correctly when
bundling server-side files that get transitively included. Removed the node:
prefix from 17 files:

- src/lib/db/migrationRunner.ts (node:fs, node:path, node:url)
- src/lib/db/core.ts (node:path, node:fs)
- src/lib/db/backup.ts (node:path, node:fs)
- src/lib/db/prompts.ts (node:fs)
- src/lib/dataPaths.ts (node:path, node:os)
- src/app/api/settings/route.ts
- src/app/api/storage/health/route.ts
- src/app/api/oauth/[provider]/[action]/route.ts
- src/app/api/db-backups/{exportAll,import,export}/route.ts
- src/shared/middleware/correlationId.ts
- src/shared/utils/requestId.ts
- src/lib/apiBridgeServer.ts
- src/lib/cacheLayer.ts
- src/lib/semanticCache.ts
- src/lib/oauth/providers/kimi-coding.ts

Also updated generate-release.md: Docker Hub sync and dual-VPS deploy
are now mandatory steps in every release.
v2.6.6 		2026-03-17 04:24:46 -03:00
diegosouzapw
a8ab16a720 chore(release): v2.6.5 — reasoning params filter, local 404 fix, Kilo Gateway, dep bumps 
- fix(sse): strip unsupported params for o1/o1-mini/o1-pro/o3/o3-mini (PR #412 @Regis-RCR)
- fix(sse): model-only lockout (5s) for local provider 404 (PR #410 @Regis-RCR)
- feat(api): Kilo Gateway provider — 335+ models, alias 'kg' (PR #408 @Regis-RCR)
- deps: better-sqlite3 12.8, undici 7.24.4, https-proxy-agent 8 (PR #413)
v2.6.5 		2026-03-17 03:05:45 -03:00
rexname
8091b6b508 feat: implement proxy registry, management APIs, docs, and test hardening 2026-03-17 13:05:27 +07:00
Diego Rodrigues de Sa e Souza
a00ef0fc7e Merge pull request #413 from diegosouzapw/dependabot/npm_and_yarn/production-4d4ff746af 
deps: bump the production group with 5 updates
 2026-03-17 03:03:49 -03:00
Diego Rodrigues de Sa e Souza
5ce6d615a4 Merge pull request #408 from Regis-RCR/feat/kilo-gateway-provider 
feat(api): add Kilo Gateway provider
 2026-03-17 03:03:47 -03:00