feat(release): v2.0.10 — CLI fingerprint UI toggle

feat: CLI fingerprint UI toggle + fix playground i18n (#223 )
- Add CLI Fingerprint Matching card in Settings > Security tab - Per-provider toggle chips (codex, claude, github, antigravity) - Emerald-themed UI with fingerprint icon and active count - Settings synced to runtime cache via API - Fix playground i18n missing from 29 non-English sidebar translations - Add cliCompatProviders to settings schema (Zod validation) - Add setCliCompatProviders/getCliCompatProviders to cliFingerprints.ts - Add i18n keys for CLI fingerprint settings (en + pt-BR)
2026-03-07 10:57:16 -03:00 · 2026-03-07 10:56:58 -03:00 · 2026-03-07 10:40:00 -03:00 · 2026-03-07 10:38:49 -03:00 · 2026-03-07 10:26:34 -03:00 · 2026-03-07 10:25:28 -03:00
112 changed files with 2584 additions and 1265 deletions
@@ -4,52 +4,73 @@ description: Deploy the latest OmniRoute code to the Akamai VPS (69.164.221.35)

 # Deploy to VPS Workflow

-Deploy OmniRoute to the production VPS using Node.js + PM2 (no Docker).
+Deploy OmniRoute to the production VPS using `npm install -g` + PM2.

 **VPS:** `69.164.221.35` (Akamai, Ubuntu 24.04, 1GB RAM + 2.5GB swap)
-**App path:** `/opt/omniroute-app`
+**Local VPS:** `192.168.0.15` (same setup)
 **Process manager:** PM2 (`omniroute`)
 **Port:** `20128`

+> [!IMPORTANT]
+> PM2 runs from the global npm package at `/usr/lib/node_modules/omniroute`.
+> **DO NOT** use git clone or local copies. The `npm install -g` command handles
+> building, publishing, and installing the standalone app in one step.
+
 ## Steps

-### 1. Push to GitHub
+### 1. Publish to npm

-Ensure all changes are committed and pushed:
+Ensure the version in `package.json` is bumped and the package is published:

 ```bash
-git push origin main
+npm publish
 ```

-### 2. SSH into VPS, pull latest code, rebuild, and restart
+### 2. Install on VPS and restart PM2

 // turbo-all

 ```bash
-ssh root@69.164.221.35 "
-  cd /opt/omniroute-app &&
-  git fetch origin &&
-  git reset --hard origin/main &&
-  export NODE_OPTIONS='--max-old-space-size=1536' &&
-  npm install --no-audit --no-fund &&
-  npm run build &&
-  pm2 restart omniroute &&
-  pm2 save &&
-  echo '✅ Deploy complete!'
-"
+ssh root@69.164.221.35 "npm install -g omniroute@latest && pm2 restart omniroute && pm2 save && echo '✅ Deploy complete!'"
+```
+
+For the local VPS:
+
+```bash
+ssh root@192.168.0.15 "npm install -g omniroute@latest && pm2 restart omniroute && pm2 save && echo '✅ Deploy complete!'"
 ```

 ### 3. Verify the deployment

 ```bash
-ssh root@69.164.221.35 "pm2 list && curl -s -o /dev/null -w 'HTTP %{http_code}' http://localhost:20128/"
+ssh root@69.164.221.35 "pm2 list && cat \$(npm root -g)/omniroute/package.json | grep version | head -1 && curl -s -o /dev/null -w 'HTTP %{http_code}' http://localhost:20128/"
 ```

-Expected: PM2 shows `online`, HTTP returns `307` (redirect to login).
+Expected: PM2 shows `online`, version matches published, HTTP returns `307` (redirect to login).
+
+## How it works
+
+1. `npm publish` builds Next.js standalone + bundles everything into the npm package
+2. `npm install -g omniroute@latest` downloads and installs to `/usr/lib/node_modules/omniroute/`
+3. PM2 is registered to run `npm start` from that directory (cwd: `/usr/lib/node_modules/omniroute`)
+4. `pm2 restart omniroute` picks up the new code immediately
+
+## PM2 Setup (one-time)
+
+If PM2 needs to be reconfigured from scratch:
+
+```bash
+ssh root@<VPS> "
+  cd /usr/lib/node_modules/omniroute &&
+  PORT=20128 pm2 start app/server.js --name omniroute --env PORT=20128 &&
+  pm2 save &&
+  pm2 startup
+"
+```

 ## Notes

- The VPS has only 1GB RAM. `NODE_OPTIONS='--max-old-space-size=1536'` uses swap for the build.
+- The `.env` file is at `/usr/lib/node_modules/omniroute/.env`. Back it up before major npm updates.
 - PM2 is configured with `pm2 startup` to auto-restart on reboot.
- The `.env` file is at `/opt/omniroute-app/.env` (copied from the old Docker setup at `/opt/omniroute/.env`).
 - Nginx proxies `omniroute.online` → `localhost:20128`.
+- The VPS has only 1GB RAM — builds happen locally via `npm publish`, not on the VPS.
@@ -31,14 +31,14 @@ jobs:
        uses: docker/setup-buildx-action@v3

      - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
        with:
          username: ${{ secrets.DOCKERHUB_USERNAME }}
          password: ${{ secrets.DOCKERHUB_TOKEN }}

      - name: Build and push by digest
        id: build
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v7
        with:
          context: .
          target: runner-base
@@ -87,7 +87,7 @@ jobs:
          merge-multiple: true

      - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
        with:
          username: ${{ secrets.DOCKERHUB_USERNAME }}
          password: ${{ secrets.DOCKERHUB_TOKEN }}
@@ -78,7 +78,7 @@ jobs:
          cache: npm

      - name: Cache node_modules
-        uses: actions/cache@v4
+        uses: actions/cache@v5
        with:
          path: node_modules
          key: ${{ runner.os }}-node-${{ hashFiles('package-lock.json') }}
@@ -120,7 +120,7 @@ jobs:
          fi

      - name: Upload artifacts
-        uses: actions/upload-artifact@v4
+        uses: actions/upload-artifact@v7
        with:
          name: electron-${{ matrix.platform }}
          path: release-assets/
@@ -136,7 +136,7 @@ jobs:
          fetch-depth: 0

      - name: Download all artifacts
-        uses: actions/download-artifact@v4
+        uses: actions/download-artifact@v8
        with:
          path: release-assets
          merge-multiple: true
@@ -172,6 +172,5 @@ jobs:
            release-assets/*.blockmap
            release-assets/*.source.tar.gz
            release-assets/*.source.zip
-            release-assets/OmniRoute.exe
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
@@ -117,3 +117,8 @@ icon.iconset/

 # VS Code Extension (independent Git repo)
 vscode-extension/
+
+# SQLite residual files
+*.sqlite-shm
+*.sqlite-wal
+*.sqlite-journal
@@ -7,6 +7,220 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0

 ---

+## [2.0.9] — 2026-03-07
+
+> ### 🚀 Feature Drop — Playground, CLI Fingerprints, ACP
+
+### ✨ New Features
+
+- **#234 — Model Playground** — Dashboard page to test any model directly (provider/model/endpoint selectors, Monaco Editor, streaming, abort, timing metrics). Available in the Debug sidebar section.
+- **#223 — CLI Fingerprint Matching** — Per-provider header/body field ordering to match native CLI binary fingerprints, reducing account flagging risk. Enable via `CLI_COMPAT_<PROVIDER>=1` or `CLI_COMPAT_ALL=1` env vars.
+- **#235 — ACP Support** — Agent Client Protocol module with CLI agent discovery (Codex, Claude, Goose, Gemini CLI, OpenClaw), process spawner/manager, and `/api/acp/agents` endpoint.
+
+### 🧹 Housekeeping
+
+- **#192 & #200** — Closed as stale (needs-info, v1.8.1, no reproduction info provided)
+
+---
+
+## [2.0.8] — 2026-03-07
+
+> ### 🐛 Bug Fix — Custom Image Model Handler Resolution
+
+### 🐛 Bug Fixes
+
+- **#238 — Custom image models still fail in handler layer** — v2.0.7 fixed the route-layer validation, but the handler (`handleImageGeneration()`) called `parseImageModel()` again internally, rejecting custom models a second time. Fix: handler now accepts an optional `resolvedProvider` parameter; when provided, it skips re-validation and routes custom models to the OpenAI-compatible handler with a synthetic config. PR #239
+
+### 📁 Files Changed
+
+| File                                         | Change                                                                           |
+| -------------------------------------------- | -------------------------------------------------------------------------------- |
+| `open-sse/handlers/imageGeneration.ts`       | Added `resolvedProvider` param + custom model fallback                           |
+| `src/app/api/v1/images/generations/route.ts` | Tracks `isCustomModel`, passes `resolvedProvider`, credentials for custom models |
+
+---
+
+## [2.0.7] — 2026-03-07
+
+> ### 🐛 Bug Fixes — Custom Image Models + Codex OAuth Workspace Isolation
+
+### 🐛 Bug Fixes
+
+- **#232 — Custom Gemini image models fail on `/v1/images/generations`** — Custom models tagged with `supportedEndpoints: ["images"]` appeared in the model listing (GET) but were rejected by the POST handler. `parseImageModel()` only checked the built-in `IMAGE_PROVIDERS` registry. Fix: added a custom model DB fallback for models with the `images` endpoint tag. PR #237
+- **#236 — Codex OAuth overwrites existing connection when same email added to another workspace** — The OAuth callback route had 3 upsert blocks matching connections by email-only, bypassing the workspace-aware logic in `createProviderConnection()`. When the same email authenticated to a new workspace, the existing connection's `workspaceId` was silently overwritten. Fix: for Codex, the match now also checks `providerSpecificData.workspaceId`, allowing separate connections per workspace. PR #237
+
+### 📁 Files Changed
+
+| File                                             | Change                                               |
+| ------------------------------------------------ | ---------------------------------------------------- |
+| `src/app/api/v1/images/generations/route.ts`     | Custom model DB fallback in POST handler             |
+| `src/app/api/oauth/[provider]/[action]/route.ts` | Workspace-aware Codex matching in 3 upsert locations |
+
+### ⏭️ Issues Triaged
+
+- **#234** — Playground feature request — Acknowledged, added to roadmap
+- **#235** — ACP support feature request — Acknowledged, added to roadmap
+
+---
+
+## [2.0.6] — 2026-03-07
+
+> ### 🐛 Bug Fix — Custom Model API Format Routing
+
+### 🐛 Bug Fixes
+
+- **#204 — Custom model `apiFormat` not used in routing** — Custom models configured with `apiFormat: "responses"` in the dashboard were still being routed through the Chat Completions translator. The `apiFormat` field was stored in the DB and displayed in the UI, but never consumed by the routing layer. Fix: `getModelInfo()` now returns `apiFormat` from the custom model DB, and both `resolveModelOrError()` functions override `targetFormat` to `openai-responses` when set. PR #233
+
+### ✅ Issues Closed
+
+- **#205** — Combo endpoint support — Already implemented in v2.0.2
+- **#206** — Manual model→endpoint mapping — Already implemented in v2.0.2
+- **#223** — CLI fingerprint parity — Responded with 4-phase roadmap
+
+### 📁 Files Changed
+
+| File                              | Change                                                                 |
+| --------------------------------- | ---------------------------------------------------------------------- |
+| `src/sse/services/model.ts`       | Added `lookupCustomModelApiFormat()`, enriched `getModelInfo()` return |
+| `src/sse/handlers/chat.ts`        | Override `targetFormat` when `apiFormat === "responses"`               |
+| `src/sse/handlers/chatHelpers.ts` | Same override in duplicate `resolveModelOrError()`                     |
+
+---
+
+## [2.0.5] — 2026-03-06
+
+> ### 🐛 Bug Fix, Electron Auto-Update & Dependency Bumps
+
+### 🐛 Bug Fixes
+
+- **#224 — Chat→Responses translation creates invalid reasoning IDs** — Removed synthetic reasoning item generation in `openaiToOpenAIResponsesRequest()`. The translator was creating reasoning items with IDs like `reasoning_15`, but OpenAI's Responses API requires server-generated `rs_*` IDs, causing `400 Invalid Request` errors from Responses-compatible upstreams. Fix: omit reasoning items entirely during translation
+- **CI: duplicate OmniRoute.exe in release workflow** — Removed redundant explicit `release-assets/OmniRoute.exe` entry that caused `softprops/action-gh-release` to fail with 404 on duplicate upload. PR #222 by @benzntech
+
+### ✨ New Features
+
+- **Electron Auto-Update** — Added auto-update functionality to the desktop app using `electron-updater`. Includes IPC handlers for check/download/install, "Check for Updates" in system tray menu, desktop notification when update is ready, and silent startup check (3s delay). PR #221 by @benzntech
+
+### 📦 Dependencies
+
+- Bump `actions/cache` from 4 to 5 (#225)
+- Bump `actions/download-artifact` from 4 to 8 (#226)
+- Bump `docker/login-action` from 3 to 4 (#227)
+- Bump `actions/upload-artifact` from 4 to 7 (#228)
+- Bump `docker/build-push-action` from 6 to 7 (#229)
+- Bump `express-rate-limit` from 8.2.1 to 8.3.0 (#230)
+
+### 📁 Files Changed
+
+| File                                              | Change                                               |
+| ------------------------------------------------- | ---------------------------------------------------- |
+| `open-sse/translator/request/openai-responses.ts` | Remove synthetic reasoning item generation           |
+| `.github/workflows/electron-release.yml`          | Remove duplicate exe entry, bump GH Actions          |
+| `.github/workflows/docker-publish.yml`            | Bump docker/login-action and build-push-action       |
+| `electron/main.js`                                | Auto-updater setup, IPC handlers, tray menu          |
+| `electron/package.json`                           | Added electron-updater dep and GitHub publish config |
+| `electron/preload.js`                             | Exposed update APIs via contextBridge                |
+| `package-lock.json`                               | Updated express-rate-limit                           |
+
+---
+
+## [2.0.4] — 2026-03-06
+
+> ### 🐛 Bug Fixes — Round-Robin Persistence & Docker Compatibility
+
+### 🐛 Bug Fixes
+
+- **#218 — Round-robin sticks to one account** — Added `last_used_at` column to `provider_connections` schema. Round-robin routing relied on `lastUsedAt` to rotate between accounts, but the column was missing from the database — the value was always `null`, causing selection to fall back to the same account. Includes auto-migration for existing databases
+- **#217 — `Cannot find module 'zod'` in Docker/standalone builds** — Added `zod` to `serverExternalPackages` in `next.config.mjs`. Next.js standalone builds weren't tracing `zod` through dynamic imports, causing crashes on Docker startup. Data is **not lost** — the crash prevented the server from reading the existing database
+
+### 📁 Files Changed
+
+| File                      | Change                                                 |
+| ------------------------- | ------------------------------------------------------ |
+| `src/lib/db/core.ts`      | Schema + migration + JSON migration for `last_used_at` |
+| `src/lib/db/providers.ts` | INSERT + UPDATE SQL for `last_used_at`                 |
+| `next.config.mjs`         | `serverExternalPackages: ['better-sqlite3', 'zod']`    |
+
+---
+
+## [2.0.3] — 2026-03-05
+
+> ### 🐛 Bug Fixes & Quota System Hardening
+
+### 🐛 Bug Fixes
+
+- **#215 — Deferred tools 400 error** — Skip `cache_control` on tools with `defer_loading=true` when assigning prompt caching to the last tool. Previously, the API rejected requests with 400 when MCP tools (Playwright, etc.) had deferred loading enabled. Fix applied in both `claudeHelper.ts` and `openai-to-claude.ts` translation layers. PR #216 by @DavyMassoneto
+- **Stale compiled schemas.js** — Deleted stale compiled `schemas.js` (912 lines) that shadowed the TypeScript `.ts` source, causing `cloudSyncActionSchema` warnings in the dashboard. PR #216 by @DavyMassoneto
+- **#202 — False quota exhaustion** — Fixed empty API responses (`{}`) creating quota objects with `utilization ?? 0` = 0% remaining, incorrectly marking accounts as exhausted. Added `hasUtilization()` guard. PR #214 by @DavyMassoneto
+- **Invalid date crash** — `parseDate()` now validates dates before comparison, handling `Invalid Date` from malformed `resetAt` values gracefully. PR #214 by @DavyMassoneto
+- **`total=0` false infinite quota** — `normalizeQuotas` now defaults to 0% remaining when `total` is zero (was incorrectly reporting 100%). PR #214 by @DavyMassoneto
+- **Tailwind v4 build failure** — Tailwind v4 scanned `*.sqlite-shm`/`.sqlite-wal` binary files, triggering "Invalid code point" errors. Added `@source not` exclusions in `globals.css`. PR #214 by @DavyMassoneto
+
+### ✨ Improvements
+
+- **Quota-aware account selection** — All load-balancing strategies (sticky, round-robin, p2c, random, least-used, cost-optimized, fill-first) now prioritize accounts with available quota over exhausted ones. PR #214 by @DavyMassoneto
+- **Concurrent refresh protection** — `tickRunning` flag prevents overlapping background quota refresh ticks; `refreshingSet` deduplicates per-connection refreshes. Thundering herd prevention with `MAX_CONCURRENT_REFRESHES=5`. PR #214 by @DavyMassoneto
+- **`clearModelUnavailability` on success** — Model unavailability is now cleared on every successful request, not only on fallback paths. PR #214 by @DavyMassoneto
+- **Centralized `anthropic-version`** — Hardcoded `anthropic-version` header (3 occurrences) centralized into `CLAUDE_CONFIG.apiVersion`. PR #214 by @DavyMassoneto
+- **Extracted `safePercentage()` utility** — Shared percentage validation function extracted to `src/shared/utils/formatting.ts`, eliminating duplication between backend and frontend. PR #214 by @DavyMassoneto
+- **`isRecord()` type guard** — Replaces inline `typeof` chain in usage API route. PR #214 by @DavyMassoneto
+
+### 📁 Files Changed
+
+| File                                                                                  | Change                                                     |
+| ------------------------------------------------------------------------------------- | ---------------------------------------------------------- |
+| `open-sse/translator/helpers/claudeHelper.ts`                                         | Skip `cache_control` on deferred tools                     |
+| `open-sse/translator/request/openai-to-claude.ts`                                     | Same fix in translator layer                               |
+| `src/shared/validation/schemas.js`                                                    | **DELETED** — stale compiled JS                            |
+| `.gitignore`                                                                          | Exclude Tailwind binary scanning                           |
+| `open-sse/services/usage.ts`                                                          | Legacy endpoint fallback logging                           |
+| `src/domain/quotaCache.ts`                                                            | **NEW** — Core quota cache with hardening                  |
+| `src/shared/utils/formatting.ts`                                                      | **NEW** — `safePercentage()` utility                       |
+| `src/instrumentation.ts`                                                              | Startup log for quota cache                                |
+| `src/sse/handlers/chat.ts`                                                            | `clearModelUnavailability` + `markAccountExhaustedFrom429` |
+| `src/sse/services/auth.ts`                                                            | Quota-aware account selection                              |
+| `src/app/globals.css`                                                                 | Tailwind `@source not` exclusions                          |
+| `src/app/api/usage/[connectionId]/route.ts`                                           | `isRecord()` type guard                                    |
+| `src/app/(dashboard)/dashboard/usage/components/ProviderLimits/ProviderLimitCard.tsx` | Use `remainingPercentage` directly                         |
+| `src/app/(dashboard)/dashboard/usage/components/ProviderLimits/utils.tsx`             | Use shared `safePercentage()`                              |
+
+---
+
+## [2.0.2] — 2026-03-05
+
+> ### 🐛 Bug Fixes & ✨ Endpoint-Aware Model Management
+
+### 🐛 Bug Fixes
+
+- **#212 — API Key creation crash** — Auto-generate `API_KEY_SECRET` at startup (like `JWT_SECRET`) to prevent HMAC crashes
+- **#213 — Circuit breaker scope** — Changed circuit breaker key from provider-level to model-level; a 429 on one account no longer blocks all accounts for the same provider
+- **#200 — Custom provider connection check** — Added connectivity fallback for OpenAI-compatible providers (Ollama, LM Studio); if `/models` and `/chat/completions` fail, a simple HTTP ping to the base URL marks the provider as connected
+
+### ✨ New Features
+
+- **#204 — API Format selector** — Custom models can now specify `apiFormat`: `chat-completions` (default) or `responses` (for the Responses API)
+- **#205 — Combo endpoint support** — Combos now accept an `endpoint` field in the schema (`chat` | `embeddings` | `images`), enabling fallback/rotation combos for non-chat endpoints
+- **#206 — Supported Endpoints mapping** — When adding custom models, users can check which endpoints the model supports (💬 Chat, 📐 Embeddings, 🖼️ Images, 🔊 Audio). Models tagged for embeddings appear in `/v1/embeddings` and models tagged for images appear in `/v1/images/generations`
+- **Visual badges** — Model rows now display colored badges for non-default API formats and endpoint types
+- **Model catalog metadata** — `/v1/models` response includes `api_format`, `type`, and `supported_endpoints` for custom models
+
+### 📁 Files Changed
+
+| File                                                    | Change                                           |
+| ------------------------------------------------------- | ------------------------------------------------ |
+| `src/instrumentation.ts`                                | Auto-generate `API_KEY_SECRET`                   |
+| `open-sse/services/combo.ts`                            | Circuit breaker keyed per-model                  |
+| `src/lib/providers/validation.ts`                       | Connectivity fallback ping                       |
+| `src/lib/db/models.ts`                                  | `apiFormat` + `supportedEndpoints` fields        |
+| `src/shared/schemas/validation.ts`                      | `endpoint` in `comboSchema`                      |
+| `src/shared/validation/schemas.ts`                      | Extended `providerModelMutationSchema`           |
+| `src/app/api/provider-models/route.ts`                  | Pass new fields through API                      |
+| `src/app/(dashboard)/dashboard/providers/[id]/page.tsx` | API format dropdown, endpoint checkboxes, badges |
+| `src/app/api/v1/models/catalog.ts`                      | Custom model metadata enrichment                 |
+| `src/app/api/v1/embeddings/route.ts`                    | Include custom embedding models                  |
+| `src/app/api/v1/images/generations/route.ts`            | Include custom image models                      |
+
+---
+
 ## [2.0.0] — 2026-03-05

 > ### 🚀 Major Release — MCP Multi-Transport, A2A Protocol, Auto-Combo Engine & Full Type Safety Overhaul
@@ -131,13 +345,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 | `src/lib/a2a/skills/`            | `smartRouting.ts`, `quotaManagement.ts`                                                                                                                    |
 | `src/app/a2a/`                   | `route.ts` (JSON-RPC 2.0 dispatch handler)                                                                                                                 |
 | `src/app/api/mcp/sse/`           | `route.ts` (SSE transport endpoint)                                                                                                                        |
-| `src/app/api/mcp/stream/`        | `route.ts` (Streamable HTTP transport endpoint)                                                                                                             |
+| `src/app/api/mcp/stream/`        | `route.ts` (Streamable HTTP transport endpoint)                                                                                                            |
 | `open-sse/services/autoCombo/`   | `scoring.ts`, `taskFitness.ts`, `engine.ts`, `selfHealing.ts`, `modePacks.ts`, `persistence.ts`, `index.ts`                                                |
 | `src/shared/contracts/`          | `quota.ts` (shared API contracts)                                                                                                                          |
 | `src/shared/constants/`          | `mcpScopes.ts`                                                                                                                                             |
 | `src/shared/validation/`         | `settingsSchemas.ts` (extracted settings Zod schema)                                                                                                       |
 | `src/lib/db/migrations/`         | `002_mcp_a2a_tables.sql`                                                                                                                                   |
-| `src/app/(dashboard)/`           | `dashboard/mcp/page.tsx`, `dashboard/a2a/page.tsx`, `dashboard/auto-combo/page.tsx`, `dashboard/endpoint/ApiEndpointsTab.tsx`                               |
+| `src/app/(dashboard)/`           | `dashboard/mcp/page.tsx`, `dashboard/a2a/page.tsx`, `dashboard/auto-combo/page.tsx`, `dashboard/endpoint/ApiEndpointsTab.tsx`                              |
 | `vscode-extension/src/services/` | `mcpClient.ts`, `a2aClient.ts`, `policyEngine.ts`, `preflightDialog.ts`, `budgetGuard.ts`, `healthMonitor.ts`, `modePackSelector.ts`, `humanCheckpoint.ts` |
 | `scripts/`                       | `check-cycles.mjs`, `check-docs-sync.mjs`, `check-route-validation.mjs`, `check-t11-any-budget.mjs`, `run-playwright-tests.mjs`, `runtime-env.mjs`         |
 | `tests/`                         | `t06-schema-hardening.test.mjs`, `t07-no-log-key-config.test.mjs`, `t08-mcp-scope-enforcement.test.mjs`, `ecosystem.test.ts`                               |
@@ -158,12 +372,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 | `889e2ba` | 2026-03-04 | feat: add error pages, harden DB layer and compliance module                             |
 | `cbd0b1c` | 2026-03-04 | refactor: harden open-sse services, eliminate any casts, add dashboard pages             |
 | `b33a853` | 2026-03-04 | feat: Introduce A2A lifecycle management, add type safety to ComfyUI and stream handling |
-| `a1a2610` | 2026-03-04 | feat: v2.0.0 - MCP server, A2A agent, proxy improvements and docs update                |
+| `a1a2610` | 2026-03-04 | feat: v2.0.0 - MCP server, A2A agent, proxy improvements and docs update                 |
 | `d615ca5` | 2026-03-05 | feat: configurable tool name prefix (#199) and custom rpm/tpm rate limits (#198)         |
 | `6d8868b` | 2026-03-05 | fix: extract validation helpers to fix webpack barrel-file resolution bug                |
 | `bc2e60c` | 2026-03-05 | feat: Introduce new A2A and MCP API routes, enhance dashboard UI, E2E tests              |
 | `79c23df` | 2026-03-05 | feat: Add i18n for media/themes, enhance combos with strategy guides, E2E tests          |
-| `2490ba5` | 2026-03-05 | feat: Introduce combo readiness checks and strategy recommendations                     |
+| `2490ba5` | 2026-03-05 | feat: Introduce combo readiness checks and strategy recommendations                      |
 | `48dda26` | 2026-03-05 | fix: CORS headers on early-return error responses + auto-combo validation (#208, #209)   |
 | `078a42b` | 2026-03-05 | feat: consolidate Endpoint, MCP, A2A into tabbed Endpoints page                          |
 | `6f1e6a0` | 2026-03-05 | feat: add MCP/A2A enable/disable toggle switches on Endpoints page                       |
@@ -287,6 +287,7 @@ Result: Never stop coding, minimal cost
 - **مكافحة الرعد القطيع** — Mutex + حماية الإشارة ضد عواصف إعادة المحاولة المتزامنة
 - **السلاسل الاحتياطية المجمعة** — إذا فشل الموفر الأساسي، فسيتم دخوله تلقائيًا عبر السلسلة دون أي تدخل
 - **Combo Circuit Breaker** — التعطيل التلقائي لمقدمي الخدمات الفاشلين ضمن سلسلة التحرير والسرد
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **لوحة معلومات الصحة** — مراقبة وقت التشغيل، وحالات قاطع الدائرة، وعمليات التأمين، وإحصائيات ذاكرة التخزين المؤقت، ووقت الاستجابة p50/p95/p99

 </details>
@@ -874,10 +875,10 @@ npm run electron:build:linux   # Linux (.AppImage)
 ### 🤖 عمليات الوكيل والبروتوكول (الإصدار 2.0)| ميزة | ماذا يفعل |

 | ------------------------------------ | -------------------------------------------------------------------------------- |
-| 🔧 **خادم MCP (16 أداة)** | تتحكم أدوات IDE/agent في التوجيه والصحة والمجموعات والحدود والعمليات | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`) 
+| 🔧 **خادم MCP (16 أداة)** | تتحكم أدوات IDE/agent في التوجيه والصحة والمجموعات والحدود والعمليات | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`)
 | 🤝 **خادم A2A (JSON-RPC + SSE)** | تنفيذ المهام من وكيل إلى وكيل مع تدفقات المزامنة والتدفق |
-| 🧭 **Consolidated Endpoints Page**   | Tabbed management page with Endpoint Proxy, MCP, A2A, and API Endpoints tabs    |
-| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF)        |
+| 🧭 **Consolidated Endpoints Page** | Tabbed management page with Endpoint Proxy, MCP, A2A, and API Endpoints tabs |
+| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF) |
 | 🛰️ **نبضات وقت تشغيل MCP** | حالة العملية الحقيقية (معرف المنتج، وقت التشغيل، عمر نبضات القلب، النقل، وضع النطاق) |
 | 📋 **مسار تدقيق MCP** | سجلات التدقيق القابلة للتصفية مع النجاح/الفشل والإسناد الرئيسي |
 | 🔐 **تنفيذ نطاق MCP** | 9 أذونات نطاق تفصيلية للوصول إلى الأدوات الخاضعة للرقابة |
@@ -287,6 +287,7 @@ OpenAI използва един формат, Claude (Anthropic) използв
 - **Anti-Thundering Herd** — Mutex + семафорна защита срещу едновременни повторни бури
 - **Combo Fallback Chains** — Ако основният доставчик се провали, автоматично преминава през веригата без намеса
 - **Combo Circuit Breaker** — Автоматично деактивира неизправните доставчици в рамките на комбинирана верига
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Health Dashboard** — Мониторинг на времето на работа, състояния на прекъсвачи, блокировки, статистика на кеша, латентност на p50/p95/p99

 </details>
@@ -875,10 +876,10 @@ OmniRoute v2.0 е създаден като операционна платфо
 ### 🤖 Операции на агент и протокол (v2.0)| Характеристика | Какво прави |

 | ------------------------------------ | -------------------------------------------------------------------------------- |
-| 🔧 **MCP сървър (16 инструмента)** | Инструментите за IDE/агент контролират маршрутизиране, здраве, комбинации, ограничения и операции | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`) 
+| 🔧 **MCP сървър (16 инструмента)** | Инструментите за IDE/агент контролират маршрутизиране, здраве, комбинации, ограничения и операции | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`)
 | 🤝 **A2A сървър (JSON-RPC + SSE)** | Изпълнение на задачи от агент към агент със синхронизиране и поточно предаване |
-| 🧭 **Consolidated Endpoints Page**   | Tabbed management page with Endpoint Proxy, MCP, A2A, and API Endpoints tabs    |
-| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF)        |
+| 🧭 **Consolidated Endpoints Page** | Tabbed management page with Endpoint Proxy, MCP, A2A, and API Endpoints tabs |
+| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF) |
 | 🛰️ **MCP Runtime Heartbeat** | Реално състояние на процеса (pid, време на работа, възраст на сърдечния ритъм, транспорт, режим на обхвата) |
 | 📋 **MCP одитна пътека** | Филтрируеми журнали за одит с успех/неуспех и ключово приписване |
 | 🔐 **Прилагане на обхват на MCP** | 9 подробни разрешения за обхват за контролиран достъп до инструменти |
@@ -283,10 +283,12 @@ Når en AI-gateway eksponeres for netværket (LAN, VPS, Docker), kan enhver med
 **Sådan løser OmniRoute det:**

 - **Circuit Breaker pr. udbyder** — Automatisk åbning/lukning med konfigurerbare tærskler og nedkøling (lukket/åbent/halvt åbent)
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Eksponentiel backoff** — Progressive forsinkelser af genforsøg
 - **Anti-tordenbesætning** — Mutex + semaforbeskyttelse mod samtidige genforsøgsstorme
 - **Combo Fallback Chains** — Hvis den primære udbyder fejler, falder den automatisk gennem kæden uden indgriben
 - **Combo Circuit Breaker** - Deaktiverer automatisk fejlende udbydere i en kombinationskæde
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Health Dashboard** — Oppetidsovervågning, strømafbrydertilstande, lockouts, cachestatistik, p50/p95/p99 latency

 </details>
@@ -875,10 +877,10 @@ OmniRoute v2.0 er bygget som en operationel platform, ikke kun en relæ-proxy.
 ### 🤖 Agent- og protokoloperationer (v2.0)| Funktion | Hvad det gør |

 | ------------------------------------ | ---------------------------------------------------------------------------------- |
-| 🔧 **MCP-server (16 værktøjer)** | IDE/agent-værktøjer kontrollerer routing, sundhed, kombinationer, grænser og operationer | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`) 
+| 🔧 **MCP-server (16 værktøjer)** | IDE/agent-værktøjer kontrollerer routing, sundhed, kombinationer, grænser og operationer | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`)
 | 🤝 **A2A-server (JSON-RPC + SSE)** | Agent-til-agent opgaveudførelse med synkronisering og streaming flows |
 | 🧭 **Consolidated Endpoints Page** | Dedikerede administrationssider (`/dashboard/mcp`, `/dashboard/a2a`) |
-| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF)        |
+| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF) |
 | 🛰️ **MCP Runtime Heartbeat** | Reel processtatus (pid, oppetid, hjerteslagsalder, transport, omfangstilstand) |
 | 📋 **MCP Audit Trail** | Filtrerbare revisionslogfiler med succes/fejl og nøgletilskrivning |
 | 🔐 **MCP Scope Enforcement** | 9 granulære omfangstilladelser til kontrolleret værktøjsadgang |
@@ -287,6 +287,7 @@ Wenn ein KI-Gateway dem Netzwerk (LAN, VPS, Docker) zugänglich gemacht wird, ka
 - **Anti-Thundering Herd** – Mutex + Semaphor-Schutz gegen gleichzeitige Wiederholungsstürme
 - **Combo-Fallback-Ketten** – Wenn der primäre Anbieter ausfällt, fällt er automatisch durch die Kette, ohne dass ein Eingreifen erforderlich ist
 - **Combo Circuit Breaker** – Deaktiviert automatisch ausgefallene Anbieter innerhalb einer Combo-Kette
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Gesundheits-Dashboard** – Betriebszeitüberwachung, Leistungsschalterzustände, Sperren, Cache-Statistiken, p50/p95/p99-Latenz

 </details>
@@ -881,10 +882,10 @@ OmniRoute v2.0 ist als Betriebsplattform konzipiert und nicht nur als Relay-Prox
 ### 🤖 Agenten- und Protokolloperationen (v2.0)| Funktion | Was es tut |

 | ------------------------------------ | -------------------------------------------------------------------------------- |
-| 🔧 **MCP-Server (16 Tools)** | IDE-/Agent-Tools steuern Routing, Zustand, Kombinationen, Grenzwerte und Vorgänge | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`) 
+| 🔧 **MCP-Server (16 Tools)** | IDE-/Agent-Tools steuern Routing, Zustand, Kombinationen, Grenzwerte und Vorgänge | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`)
 | 🤝 **A2A-Server (JSON-RPC + SSE)** | Ausführung von Agent-zu-Agent-Aufgaben mit Synchronisierungs- und Streaming-Flows |
 | 🧭 **Consolidated Endpoints Page** | Dedizierte Verwaltungsseiten (`/dashboard/mcp`, `/dashboard/a2a`) |
-| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF)        |
+| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF) |
 | 🛰️ **MCP Runtime Heartbeat** | Echter Prozessstatus (PID, Betriebszeit, Heartbeat-Alter, Transport, Scope-Modus) |
 | 📋 **MCP Audit Trail** | Filterbare Audit-Protokolle mit Erfolg/Misserfolg und Schlüsselzuordnung |
 | 🔐 **Durchsetzung des MCP-Geltungsbereichs** | 9 granulare Umfangsberechtigungen für kontrollierten Werkzeugzugriff |
@@ -868,6 +868,7 @@ npm run electron:build:linux   # Linux (.AppImage)
 | Característica                     | Qué Hace                                                                     |
 | ---------------------------------- | ---------------------------------------------------------------------------- |
 | 🔌 **Circuit Breaker**             | Auto-apertura/cierre por proveedor con umbrales configurables                |
+| 🎯 **Endpoint-Aware Models**       | Custom models declare supported endpoints + API format                       |
 | 🛡️ **Anti-Thundering Herd**        | Mutex + semáforo rate-limit para proveedores con API key                     |
 | 🧠 **Caché Semántico**             | Caché de dos niveles (firma + semántico) reduce costo y latencia             |
 | ⚡ **Idempotencia de Solicitud**   | Ventana de dedup de 5s para solicitudes duplicadas                           |
@@ -294,6 +294,7 @@ Tekoälypalveluntarjoajat voivat muuttua epävakaiksi, palauttaa 5xx-virheitä t
 - **Anti-Thundering Herd** — Mutex + semaforisuoja samanaikaisia myrskyjä vastaan
 - **Yhdistelmävaraketjut** – Jos ensisijainen toimittaja epäonnistuu, putoaa automaattisesti ketjun läpi ilman väliintuloa
 - **Combo Circuit Breaker** — Poistaa automaattisesti käytöstä vialliset palveluntarjoajat yhdistelmäketjussa
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Health Dashboard** — käytettävyyden valvonta, katkaisijoiden tilat, lukitukset, välimuistitilastot, p50/p95/p99-viive

 </details>
@@ -866,6 +866,7 @@ npm run electron:build:linux   # Linux (.AppImage)
 | Fonctionnalité                  | Ce qu'elle fait                                                              |
 | ------------------------------- | ---------------------------------------------------------------------------- |
 | 🔌 **Circuit Breaker**          | Ouverture/fermeture auto par fournisseur avec seuils configurables           |
+| 🎯 **Endpoint-Aware Models**    | Custom models declare supported endpoints + API format                       |
 | 🛡️ **Anti-Thundering Herd**     | Mutex + sémaphore de rate-limit pour les fournisseurs avec clé API           |
 | 🧠 **Cache sémantique**         | Cache à deux niveaux (signature + sémantique) réduit coût et latence         |
 | ⚡ **Idempotence des requêtes** | Fenêtre de dédup 5s pour les requêtes dupliquées                             |
@@ -1495,14 +1495,15 @@ Se não quiser criar credenciais próprias agora, ainda é possível usar o flux

 ל-OmniRoute יש **210+ תכונות מתוכננות** לאורך שלבי פיתוח מרובים. להלן תחומי המפתח:

-| קטגוריה               | תכונות מתוכננות | הבהרה                                                                                    |
-| --------------------- | --------------- | ---------------------------------------------------------------------------------------- |
-| 🧠 **ניתוב ומודיעין** | 25+             | ניתוב עם זמן האחזור הנמוך ביותר, ניתוב מבוסס תגים, בדיקה מוקדמת של מכסה, בחירת חשבון P2C |
-| 🔒 **אבטחה ותאימות**  | 20+             | הקשחת SSRF, הסוואה של אישורים, הגבלת קצב לכל נקודת קצה, היקף מפתח ניהול                  |
-| 📊 **צפיות**          | 15+             | אינטגרציה של OpenTelemetry, ניטור מכסות בזמן אמת, מעקב עלויות לכל דגם                    |
-| 🔄 **שילובי ספקים**   | 20+             | רישום מודלים דינמיים, צינון ספקים, Codex מרובה חשבונות, ניתוח מכסת Copilot               |
-| ⚡ **ביצועים**        | 15+             | שכבת מטמון כפולה, מטמון הנחיה, מטמון תגובה, סטרימינג Keepalive, API אצווה                |
-| 🌐 **מערכת אקולוגית** | 10+             | WebSocket API, טעינה חוזרת של תצורה, חנות תצורה מבוזרת, מצב מסחרי                        |
+| קטגוריה                      | תכונות מתוכננות                                        | הבהרה                                                                                    |
+| ---------------------------- | ------------------------------------------------------ | ---------------------------------------------------------------------------------------- |
+| 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
+| 🧠 **ניתוב ומודיעין**        | 25+                                                    | ניתוב עם זמן האחזור הנמוך ביותר, ניתוב מבוסס תגים, בדיקה מוקדמת של מכסה, בחירת חשבון P2C |
+| 🔒 **אבטחה ותאימות**         | 20+                                                    | הקשחת SSRF, הסוואה של אישורים, הגבלת קצב לכל נקודת קצה, היקף מפתח ניהול                  |
+| 📊 **צפיות**                 | 15+                                                    | אינטגרציה של OpenTelemetry, ניטור מכסות בזמן אמת, מעקב עלויות לכל דגם                    |
+| 🔄 **שילובי ספקים**          | 20+                                                    | רישום מודלים דינמיים, צינון ספקים, Codex מרובה חשבונות, ניתוח מכסת Copilot               |
+| ⚡ **ביצועים**               | 15+                                                    | שכבת מטמון כפולה, מטמון הנחיה, מטמון תגובה, סטרימינג Keepalive, API אצווה                |
+| 🌐 **מערכת אקולוגית**        | 10+                                                    | WebSocket API, טעינה חוזרת של תצורה, חנות תצורה מבוזרת, מצב מסחרי                        |

 ### 🔜 בקרוב

@@ -294,6 +294,7 @@ Az AI-szolgáltatók instabillá válhatnak, 5xx-es hibákat adnak vissza, vagy
 - **Mennydörgés elleni csorda** - Mutex + szemafor védelem az egyidejű újrapróbálkozási viharok ellen
 - **Kombinált tartalék láncok** – Ha az elsődleges szolgáltató meghibásodik, automatikusan, beavatkozás nélkül átesik a láncon
 - **Combo Circuit Breaker** – Automatikusan letiltja a hibás szolgáltatókat a kombinált láncon belül
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Egészségügyi irányítópult** — Üzemidő-figyelés, áramkör-megszakító állapotok, zárolások, gyorsítótár-statisztika, p50/p95/p99 késleltetés

 </details>
@@ -294,6 +294,7 @@ Penyedia AI bisa menjadi tidak stabil, menampilkan kesalahan 5xx, atau mencapai
 - **Kawanan Anti-Guntur** — Perlindungan mutex + semaphore terhadap badai percobaan ulang secara bersamaan
 - **Combo Fallback Chains** — Jika penyedia utama gagal, otomatis gagal dalam rantai tanpa intervensi
 - **Combo Circuit Breaker** — Menonaktifkan secara otomatis penyedia yang gagal dalam rantai kombo
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Dasbor Kesehatan** — Pemantauan waktu aktif, status pemutus sirkuit, penguncian, statistik cache, latensi p50/p95/p99

 </details>
@@ -1138,14 +1138,15 @@ Cost: $0 forever!

 ओम्निरूट ने कई विकास चरणों में **210+ सुविधाओं की योजना बनाई है**। यहां प्रमुख क्षेत्र हैं:

-| श्रेणी                      | नियोजित विशेषताएं | हाइलाइट्स                                                                                   |
-| --------------------------- | ----------------- | ------------------------------------------------------------------------------------------- |
-| 🧠 **रूटिंग और इंटेलिजेंस** | 25+               | न्यूनतम-विलंबता रूटिंग, टैग-आधारित रूटिंग, कोटा प्रीफ़्लाइट, पी2सी खाता चयन                 |
-| 🔒 **सुरक्षा एवं अनुपालन**  | 20+               | एसएसआरएफ हार्डनिंग, क्रेडेंशियल क्लोकिंग, प्रति समापन बिंदु दर-सीमा, प्रबंधन कुंजी स्कोपिंग |
-| 📊 **अवलोकनशीलता**          | 15+               | ओपन टेलीमेट्री एकीकरण, वास्तविक समय कोटा निगरानी, प्रति मॉडल लागत ट्रैकिंग                |
-| 🔄 **प्रदाता एकीकरण**       | 20+               | डायनेमिक मॉडल रजिस्ट्री, प्रदाता कूलडाउन, मल्टी-अकाउंट कोडेक्स, कोपायलट कोटा पार्सिंग       |
-| ⚡ **प्रदर्शन**             | 15+               | दोहरी कैश परत, शीघ्र कैश, प्रतिक्रिया कैश, स्ट्रीमिंग कीपलाइव, बैच एपीआई                    |
-| 🌐 **पारिस्थितिकी तंत्र**   | 10+               | वेबसॉकेट एपीआई, कॉन्फिग हॉट-रीलोड, वितरित कॉन्फिग स्टोर, वाणिज्यिक मोड                      |
+| श्रेणी                       | नियोजित विशेषताएं                                      | हाइलाइट्स                                                                                   |
+| ---------------------------- | ------------------------------------------------------ | ------------------------------------------------------------------------------------------- |
+| 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
+| 🧠 **रूटिंग और इंटेलिजेंस**  | 25+                                                    | न्यूनतम-विलंबता रूटिंग, टैग-आधारित रूटिंग, कोटा प्रीफ़्लाइट, पी2सी खाता चयन                 |
+| 🔒 **सुरक्षा एवं अनुपालन**   | 20+                                                    | एसएसआरएफ हार्डनिंग, क्रेडेंशियल क्लोकिंग, प्रति समापन बिंदु दर-सीमा, प्रबंधन कुंजी स्कोपिंग |
+| 📊 **अवलोकनशीलता**           | 15+                                                    | ओपन टेलीमेट्री एकीकरण, वास्तविक समय कोटा निगरानी, प्रति मॉडल लागत ट्रैकिंग                |
+| 🔄 **प्रदाता एकीकरण**        | 20+                                                    | डायनेमिक मॉडल रजिस्ट्री, प्रदाता कूलडाउन, मल्टी-अकाउंट कोडेक्स, कोपायलट कोटा पार्सिंग       |
+| ⚡ **प्रदर्शन**              | 15+                                                    | दोहरी कैश परत, शीघ्र कैश, प्रतिक्रिया कैश, स्ट्रीमिंग कीपलाइव, बैच एपीआई                    |
+| 🌐 **पारिस्थितिकी तंत्र**    | 10+                                                    | वेबसॉकेट एपीआई, कॉन्फिग हॉट-रीलोड, वितरित कॉन्फिग स्टोर, वाणिज्यिक मोड                      |

 ### 🔜 जल्द आ रहा है

@@ -865,6 +865,7 @@ npm run electron:build:linux   # Linux (.AppImage)
 | Funzionalità                    | Cosa Fa                                                                      |
 | ------------------------------- | ---------------------------------------------------------------------------- |
 | 🔌 **Circuit Breaker**          | Apertura/chiusura auto per provider con soglie configurabili                 |
+| 🎯 **Endpoint-Aware Models**    | Custom models declare supported endpoints + API format                       |
 | 🛡️ **Anti-Thundering Herd**     | Mutex + semaforo rate-limit per provider con API key                         |
 | 🧠 **Cache semantica**          | Cache a due livelli (firma + semantica) riduce costi e latenza               |
 | ⚡ **Idempotenza richieste**    | Finestra dedup 5s per richieste duplicate                                    |
@@ -532,6 +532,7 @@ AIは単なるチャット補完ではありません。開発者は、画像の
 - MCP ダッシュボードからコンボのアクティブ化を直接切り替えます
 - 事前定義されたポリシーパックから復元プロファイルを適用
 - 同じ操作パネルからサーキットブレーカーの状態をリセット
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |

 </details>

@@ -864,6 +865,7 @@ npm run electron:build:linux   # Linux (.AppImage)
 | 特集                             | 何をするのか                                                                 |
 | -------------------------------- | ---------------------------------------------------------------------------- |
 | 🔌 **サーキットブレーカー**      | 設定可能なしきい値によるプロバイダーごとの自動開閉                           |
+| 🎯 **Endpoint-Aware Models**     | Custom models declare supported endpoints + API format                       |
 | 🛡️ **対雷鳴の群れ**              | API キープロバイダーのミューテックス + セマフォのレート制限                  |
 | 🧠 **セマンティック キャッシュ** | 2 層キャッシュ (シグネチャ + セマンティック) によりコストと遅延が削減        |
 | ⚡ **冪等性のリクエスト**        | 重複リクエストに対する 5 秒の重複除去ウィンドウ                              |
@@ -964,6 +966,7 @@ Combo: "my-coding-stack"

 - システムステータス (稼働時間、バージョン、メモリ使用量)
 - プロバイダーごとのサーキットブレーカーの状態 (クローズ/オープン/ハーフオープン)
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - レート制限ステータスとアクティブなロックアウト
 - 署名キャッシュ統計
 - レイテンシ テレメトリ (p50/p95/p99) + プロンプト キャッシュ
@@ -1495,14 +1495,15 @@ Se não quiser criar credenciais próprias agora, ainda é possível usar o flux

 OmniRoute에는 여러 개발 단계에 걸쳐 **210개 이상의 기능이 계획되어 있습니다**. 주요 영역은 다음과 같습니다.

-| 카테고리                    | 계획된 기능 | 하이라이트                                                                     |
-| --------------------------- | ----------- | ------------------------------------------------------------------------------ |
-| 🧠 **라우팅 및 인텔리전스** | 25세 이상   | 최저 대기 시간 라우팅, 태그 기반 라우팅, 실행 전 할당량, P2C 계정 선택         |
-| 🔒 **보안 및 규정 준수**    | 20세 이상   | SSRF 강화, 자격 증명 클로킹, 엔드포인트당 속도 제한, 관리 키 범위 지정         |
-| 📊 **관측성**               | 15세 이상   | OpenTelemetry 통합, 실시간 할당량 모니터링, 모델별 비용 추적                   |
-| 🔄 **공급자 통합**          | 20세 이상   | 동적 모델 레지스트리, 공급자 쿨다운, 다중 계정 Codex, Copilot 할당량 구문 분석 |
-| ⚡ **성능**                 | 15세 이상   | 듀얼 캐시 레이어, 프롬프트 캐시, 응답 캐시, 스트리밍 Keepalive, 배치 API       |
-| 🌐 **생태계**               | 10세 이상   | WebSocket API, 구성 핫 리로드, 분산 구성 저장소, 상용 모드                     |
+| 카테고리                     | 계획된 기능                                            | 하이라이트                                                                     |
+| ---------------------------- | ------------------------------------------------------ | ------------------------------------------------------------------------------ |
+| 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
+| 🧠 **라우팅 및 인텔리전스**  | 25세 이상                                              | 최저 대기 시간 라우팅, 태그 기반 라우팅, 실행 전 할당량, P2C 계정 선택         |
+| 🔒 **보안 및 규정 준수**     | 20세 이상                                              | SSRF 강화, 자격 증명 클로킹, 엔드포인트당 속도 제한, 관리 키 범위 지정         |
+| 📊 **관측성**                | 15세 이상                                              | OpenTelemetry 통합, 실시간 할당량 모니터링, 모델별 비용 추적                   |
+| 🔄 **공급자 통합**           | 20세 이상                                              | 동적 모델 레지스트리, 공급자 쿨다운, 다중 계정 Codex, Copilot 할당량 구문 분석 |
+| ⚡ **성능**                  | 15세 이상                                              | 듀얼 캐시 레이어, 프롬프트 캐시, 응답 캐시, 스트리밍 Keepalive, 배치 API       |
+| 🌐 **생태계**                | 10세 이상                                              | WebSocket API, 구성 핫 리로드, 분산 구성 저장소, 상용 모드                     |

 ### 🔜 출시 예정

@@ -47,7 +47,7 @@ _Your universal API proxy — one endpoint, 36+ providers, zero downtime. Now wi
 | **Settings**   | ![Settings](docs/screenshots/06-settings.png)     |
 | **CLI Tools**  | ![CLI Tools](docs/screenshots/07-cli-tools.png)   |
 | **Usage Logs** | ![Usage](docs/screenshots/08-usage.png)           |
-| **Endpoints**  | ![Endpoints](docs/screenshots/09-endpoint.png)   |
+| **Endpoints**  | ![Endpoints](docs/screenshots/09-endpoint.png)    |

 </details>

@@ -289,7 +289,7 @@ AI providers can become unstable, return 5xx errors, or hit temporary rate limit

 **How OmniRoute solves it:**

- **Circuit Breaker per-provider** — Auto-open/close with configurable thresholds and cooldown (Closed/Open/Half-Open)
+- **Circuit Breaker per-model** — Auto-open/close with configurable thresholds and cooldown (Closed/Open/Half-Open), scoped per-model to avoid cascading blocks
 - **Exponential Backoff** — Progressive retry delays
 - **Anti-Thundering Herd** — Mutex + semaphore protection against concurrent retry storms
 - **Combo Fallback Chains** — If the primary provider fails, automatically falls through the chain with no intervention
@@ -890,19 +890,19 @@ OmniRoute v2.0 is built as an operational platform, not just a relay proxy.

 ### 🤖 Agent & Protocol Operations (v2.0)

-| Feature                              | What It Does                                                                     |
-| ------------------------------------ | -------------------------------------------------------------------------------- |
-| 🔧 **MCP Server (16 tools)**         | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`) |
-| 🤝 **A2A Server (JSON-RPC + SSE)**   | Agent-to-agent task execution with sync and streaming flows                      |
-| 🧭 **Consolidated Endpoints Page**   | Tabbed management page with Endpoint Proxy, MCP, A2A, and API Endpoints tabs    |
-| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF)        |
-| 🛰️ **MCP Runtime Heartbeat**         | Real process status (pid, uptime, heartbeat age, transport, scope mode)          |
-| 📋 **MCP Audit Trail**               | Filterable audit logs with success/failure and key attribution                   |
-| 🔐 **MCP Scope Enforcement**         | 9 granular scope permissions for controlled tool access                          |
-| 📡 **A2A Task Lifecycle Management** | List/filter tasks, inspect events/artifacts, cancel running tasks                |
-| 📋 **Agent Card Discovery**          | `/.well-known/agent.json` for client auto-discovery                              |
-| 🧪 **Protocol E2E Test Harness**     | Real MCP SDK + A2A client flows in `test:protocols:e2e`                          |
-| ⚙️ **Operational Controls**          | Switch combo, apply resilience profiles, reset breakers from one control surface |
+| Feature                               | What It Does                                                                                       |
+| ------------------------------------- | -------------------------------------------------------------------------------------------------- |
+| 🔧 **MCP Server (16 tools)**          | IDE/agent tools via 3 transports: stdio, SSE (`/api/mcp/sse`), Streamable HTTP (`/api/mcp/stream`) |
+| 🤝 **A2A Server (JSON-RPC + SSE)**    | Agent-to-agent task execution with sync and streaming flows                                        |
+| 🧭 **Consolidated Endpoints Page**    | Tabbed management page with Endpoint Proxy, MCP, A2A, and API Endpoints tabs                       |
+| 🎚️ **Service Enable/Disable Toggles** | ON/OFF switches for MCP and A2A with settings persistence (default: OFF)                           |
+| 🛰️ **MCP Runtime Heartbeat**          | Real process status (pid, uptime, heartbeat age, transport, scope mode)                            |
+| 📋 **MCP Audit Trail**                | Filterable audit logs with success/failure and key attribution                                     |
+| 🔐 **MCP Scope Enforcement**          | 9 granular scope permissions for controlled tool access                                            |
+| 📡 **A2A Task Lifecycle Management**  | List/filter tasks, inspect events/artifacts, cancel running tasks                                  |
+| 📋 **Agent Card Discovery**           | `/.well-known/agent.json` for client auto-discovery                                                |
+| 🧪 **Protocol E2E Test Harness**      | Real MCP SDK + A2A client flows in `test:protocols:e2e`                                            |
+| ⚙️ **Operational Controls**           | Switch combo, apply resilience profiles, reset breakers from one control surface                   |

 ### 🧠 Routing & Intelligence

@@ -938,7 +938,8 @@ OmniRoute v2.0 is built as an operational platform, not just a relay proxy.

 | Feature                             | What It Does                                               |
 | ----------------------------------- | ---------------------------------------------------------- |
-| 🔌 **Circuit Breakers**             | Provider-level trip/recover with threshold controls        |
+| 🔌 **Circuit Breakers**             | Per-model trip/recover with threshold controls             |
+| 🎯 **Endpoint-Aware Models**        | Custom models declare supported endpoints + API format     |
 | 🛡️ **Anti-Thundering Herd**         | Mutex + semaphore protections on retry/rate events         |
 | 🧠 **Semantic + Signature Cache**   | Cost/latency reduction with two cache layers               |
 | ⚡ **Request Idempotency**          | Duplicate protection window                                |
@@ -1495,14 +1495,15 @@ Jika anda ingin mendapatkan credenciais próprias agora, ada kemungkinan penggun

 OmniRoute mempunyai **210+ ciri yang dirancang** merentas berbilang fasa pembangunan. Berikut adalah bidang utama:

-| Kategori                       | Ciri Terancang | Sorotan                                                                                              |
-| ------------------------------ | -------------- | ---------------------------------------------------------------------------------------------------- |
-| 🧠 **Penghalaan & Perisikan**  | 25+            | Penghalaan kependaman terendah, penghalaan berasaskan teg, kuota prapenerbangan, pemilihan akaun P2C |
-| 🔒 **Keselamatan & Pematuhan** | 20+            | Pengerasan SSRF, penyelubungan kelayakan, had kadar setiap titik akhir, skop kunci pengurusan        |
-| 📊 **Kebolehlihatan**          | 15+            | Penyepaduan OpenTelemetry, pemantauan kuota masa nyata, penjejakan kos setiap model                  |
-| 🔄 **Integrasi Pembekal**      | 20+            | Pendaftaran model dinamik, penyejukan pembekal, Codex berbilang akaun, penghuraian kuota Copilot     |
-| ⚡ **Prestasi**                | 15+            | Lapisan cache dwi, cache gesaan, cache respons, penstriman keepalive, API kelompok                 |
-| 🌐 **Ekosistem**               | 10+            | API WebSocket, konfigurasi hot-reload, kedai konfigurasi teragih, mod komersial                      |
+| Kategori                       | Ciri Terancang                                         | Sorotan                                                                                              |
+| ------------------------------ | ------------------------------------------------------ | ---------------------------------------------------------------------------------------------------- |
+| 🎯 **Endpoint-Aware Models**   | Custom models declare supported endpoints + API format |
+| 🧠 **Penghalaan & Perisikan**  | 25+                                                    | Penghalaan kependaman terendah, penghalaan berasaskan teg, kuota prapenerbangan, pemilihan akaun P2C |
+| 🔒 **Keselamatan & Pematuhan** | 20+                                                    | Pengerasan SSRF, penyelubungan kelayakan, had kadar setiap titik akhir, skop kunci pengurusan        |
+| 📊 **Kebolehlihatan**          | 15+                                                    | Penyepaduan OpenTelemetry, pemantauan kuota masa nyata, penjejakan kos setiap model                  |
+| 🔄 **Integrasi Pembekal**      | 20+                                                    | Pendaftaran model dinamik, penyejukan pembekal, Codex berbilang akaun, penghuraian kuota Copilot     |
+| ⚡ **Prestasi**                | 15+                                                    | Lapisan cache dwi, cache gesaan, cache respons, penstriman keepalive, API kelompok                 |
+| 🌐 **Ekosistem**               | 10+                                                    | API WebSocket, konfigurasi hot-reload, kedai konfigurasi teragih, mod komersial                      |

 ### 🔜 Akan Datang

@@ -1495,14 +1495,15 @@ Als u geen geloofwaardige geloofwaardigheid meer heeft, is het mogelijk om de st

 OmniRoute heeft **210+ functies gepland** over meerdere ontwikkelingsfasen. Dit zijn de belangrijkste gebieden:

-| Categorie                       | Geplande functies | Hoogtepunten                                                                                              |
-| ------------------------------- | ----------------- | --------------------------------------------------------------------------------------------------------- |
-| 🧠 **Routing en intelligentie** | 25+               | Routering met de laagste latentie, op tags gebaseerde routering, quota-preflight, P2C-accountselectie     |
-| 🔒 **Beveiliging en naleving**  | 20+               | SSRF-verharding, cloaking van inloggegevens, snelheidslimiet per eindpunt, scoping van beheersleutels     |
-| 📊 **Waarneembaarheid**         | 15+               | OpenTelemetry-integratie, realtime quotabewaking, kostenregistratie per model                             |
-| 🔄 **Provider-integraties**     | 20+               | Dynamisch modelregister, cooldowns van providers, Codex met meerdere accounts, parseren van Copilot-quota |
-| ⚡ **Prestaties**               | 15+               | Dubbele cachelaag, promptcache, responscache, streaming keepalive, batch-API                              |
-| 🌐 **Ecosysteem**               | 10+               | WebSocket API, configuratie hot-reload, gedistribueerde configuratieopslag, commerciële modus             |
+| Categorie                       | Geplande functies                                      | Hoogtepunten                                                                                              |
+| ------------------------------- | ------------------------------------------------------ | --------------------------------------------------------------------------------------------------------- |
+| 🎯 **Endpoint-Aware Models**    | Custom models declare supported endpoints + API format |
+| 🧠 **Routing en intelligentie** | 25+                                                    | Routering met de laagste latentie, op tags gebaseerde routering, quota-preflight, P2C-accountselectie     |
+| 🔒 **Beveiliging en naleving**  | 20+                                                    | SSRF-verharding, cloaking van inloggegevens, snelheidslimiet per eindpunt, scoping van beheersleutels     |
+| 📊 **Waarneembaarheid**         | 15+                                                    | OpenTelemetry-integratie, realtime quotabewaking, kostenregistratie per model                             |
+| 🔄 **Provider-integraties**     | 20+                                                    | Dynamisch modelregister, cooldowns van providers, Codex met meerdere accounts, parseren van Copilot-quota |
+| ⚡ **Prestaties**               | 15+                                                    | Dubbele cachelaag, promptcache, responscache, streaming keepalive, batch-API                              |
+| 🌐 **Ecosysteem**               | 10+                                                    | WebSocket API, configuratie hot-reload, gedistribueerde configuratieopslag, commerciële modus             |

 ### 🔜 Binnenkort beschikbaar

@@ -864,6 +864,7 @@ npm run electron:build:linux   # Linux (.AppImage)
 | Funksjon                          | Hva det gjør                                                                  |
 | --------------------------------- | ----------------------------------------------------------------------------- |
 | 🔌 **Circuit Breaker**            | Automatisk åpning/lukking per leverandør med konfigurerbare terskler          |
+| 🎯 **Endpoint-Aware Models**      | Custom models declare supported endpoints + API format                        |
 | 🛡️ **Anti-tordenflokk**           | Mutex + semaforhastighetsgrense for API-nøkkelleverandører                    |
 | 🧠 **Semantisk cache**            | To-lags cache (signatur + semantisk) reduserer kostnader og ventetid          |
 | ⚡ **Be om idempotens**           | 5s dedup-vindu for dupliserte forespørsler                                    |
@@ -289,7 +289,7 @@ Ang mga tagapagbigay ng AI ay maaaring maging hindi matatag, magbalik ng 5xx na

 **Paano ito niresolba ng OmniRoute:**

- **Circuit Breaker per-provider** — Awtomatikong buksan/sarado na may mga na-configure na threshold at cooldown (Sarado/Bukas/Kalahating Bukas)
+- **Circuit Breaker per-model** — Awtomatikong buksan/sarado na may mga na-configure na threshold at cooldown (Sarado/Bukas/Kalahating Bukas)
 - **Exponential Backoff** — Progressive retry delays
 - **Anti-Thundering Herd** — Mutex + semaphore na proteksyon laban sa kasabay na muling pagsubok na mga bagyo
 - **Combo Fallback Chains** — Kung nabigo ang pangunahing provider, awtomatikong mahuhulog sa chain nang walang interbensyon
@@ -864,6 +864,7 @@ npm run electron:build:linux   # Linux (.AppImage)
 | Tampok                                    | Ano ang Ginagawa Nito                                                                 |
 | ----------------------------------------- | ------------------------------------------------------------------------------------- |
 | 🔌 **Circuit Breaker**                    | Awtomatikong buksan/isara ang bawat provider na may mga na-configure na threshold     |
+| 🎯 **Endpoint-Aware Models**              | Custom models declare supported endpoints + API format                                |
 | 🛡️ **Anti-Thundering Herd**               | Mutex + semaphore rate-limit para sa mga API key provider                             |
 | 🧠 **Semantic Cache**                     | Binabawasan ng two-tier na cache (pirma + semantiko) ang gastos at latency            |
 | ⚡ **Humiling ng Idempotency**            | 5s dedup window para sa mga duplicate na kahilingan                                   |
@@ -1495,14 +1495,15 @@ Jeśli chcesz uzyskać dostęp do **podręcznika URL**:

 OmniRoute ma **ponad 210 funkcji zaplanowanych** w wielu fazach rozwoju. Oto kluczowe obszary:

-| Kategoria                        | Planowane funkcje | Najważniejsze                                                                                                                                      |
-| -------------------------------- | ----------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- |
-| 🧠 **Routing i inteligencja**    | 25+               | Routing z najmniejszym opóźnieniem, routing oparty na tagach, wstępna inspekcja przydziału, wybór konta P2C                                        |
-| 🔒 **Bezpieczeństwo i zgodność** | 20+               | Wzmocnienie SSRF, maskowanie poświadczeń, limit szybkości na punkt końcowy, zakres kluczy zarządzania                                              |
-| 📊 **Obserwowalność**            | 15+               | Integracja OpenTelemetry, monitorowanie kwot w czasie rzeczywistym, śledzenie kosztów według modelu                                                |
-| 🔄 **Integracja dostawców**      | 20+               | Rejestr modeli dynamicznych, czasy odnowienia dostawcy, Kodeks dla wielu kont, analiza przydziału Copilot                                          |
-| ⚡ **Wydajność**                 | 15+               | Podwójna warstwa pamięci podręcznej, pamięć podręczna podpowiedzi, pamięć podręczna odpowiedzi, utrzymywanie transmisji strumieniowej, wsadowe API |
-| 🌐 **Ekosystem**                 | 10+               | WebSocket API, ładowanie konfiguracji na gorąco, rozproszony magazyn konfiguracji, tryb komercyjny                                                 |
+| Kategoria                        | Planowane funkcje                                      | Najważniejsze                                                                                                                                      |
+| -------------------------------- | ------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------- |
+| 🎯 **Endpoint-Aware Models**     | Custom models declare supported endpoints + API format |
+| 🧠 **Routing i inteligencja**    | 25+                                                    | Routing z najmniejszym opóźnieniem, routing oparty na tagach, wstępna inspekcja przydziału, wybór konta P2C                                        |
+| 🔒 **Bezpieczeństwo i zgodność** | 20+                                                    | Wzmocnienie SSRF, maskowanie poświadczeń, limit szybkości na punkt końcowy, zakres kluczy zarządzania                                              |
+| 📊 **Obserwowalność**            | 15+                                                    | Integracja OpenTelemetry, monitorowanie kwot w czasie rzeczywistym, śledzenie kosztów według modelu                                                |
+| 🔄 **Integracja dostawców**      | 20+                                                    | Rejestr modeli dynamicznych, czasy odnowienia dostawcy, Kodeks dla wielu kont, analiza przydziału Copilot                                          |
+| ⚡ **Wydajność**                 | 15+                                                    | Podwójna warstwa pamięci podręcznej, pamięć podręczna podpowiedzi, pamięć podręczna odpowiedzi, utrzymywanie transmisji strumieniowej, wsadowe API |
+| 🌐 **Ekosystem**                 | 10+                                                    | WebSocket API, ładowanie konfiguracji na gorąco, rozproszony magazyn konfiguracji, tryb komercyjny                                                 |

 ### 🔜 Już wkrótce

@@ -289,7 +289,7 @@ Provedores de IA podem ficar instáveis, retornar erro 5xx ou atingir limites te

 **Como o OmniRoute resolve isso:**

- **Circuit Breaker por provedor** — Abre/fecha automaticamente com limiares e cooldown configuráveis (Closed/Open/Half-Open)
+- **Circuit Breaker por modelo** — Abre/fecha automaticamente com limiares e cooldown configuráveis (Closed/Open/Half-Open)
 - **Exponential Backoff** — Atrasos progressivos de retry
 - **Anti-Thundering Herd** — Proteção com mutex + semáforo contra tempestade de retries concorrentes
 - **Cadeias de Fallback em Combo** — Se o primário falhar, avança automaticamente na cadeia sem intervenção
@@ -871,7 +871,8 @@ Por que isso é relevante:

 | Funcionalidade                      | O que Faz                                                                     |
 | ----------------------------------- | ----------------------------------------------------------------------------- |
-| 🔌 **Circuit Breaker**              | Auto-abertura/fechamento por provedor com limites configuráveis               |
+| 🔌 **Circuit Breaker**              | Trip/recover por modelo com limites configuráveis                             |
+| 🎯 **Endpoint-Aware Models**        | Custom models declare supported endpoints + API format                        |
 | 🛡️ **Anti-Thundering Herd**         | Mutex + semáforo rate-limit para provedores com API key                       |
 | 🧠 **Cache Semântico**              | Cache de duas camadas (assinatura + semântico) reduz custo e latência         |
 | ⚡ **Idempotência de Requisição**   | Janela de dedup de 5s para requisições duplicadas                             |
@@ -294,6 +294,7 @@ Os provedores de IA podem ficar instáveis, retornar erros 5xx ou atingir limite
 - **Rebanho Anti-Trovão** — Proteção Mutex + semáforo contra tempestades de novas tentativas simultâneas
 - **Combo Fallback Chains** — Se o provedor primário falhar, ele cairá automaticamente na cadeia sem intervenção
 - **Combo Circuit Breaker** — Desativa automaticamente provedores com falha em uma cadeia de combinação
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Health Dashboard** — Monitoramento de tempo de atividade, estados de disjuntores, bloqueios, estatísticas de cache, latência p50/p95/p99

 </details>
@@ -290,10 +290,12 @@ Furnizorii de AI pot deveni instabili, pot returna erori 5xx sau pot atinge limi
 **Cum o rezolvă OmniRoute:**

 - **Circuit Breaker per furnizor** - Deschidere/închidere automată cu praguri configurabile și răcire (Închis/Deschis/Pe jumătate deschis)
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Backoff exponențial** — Întârzieri progresive ale reîncercării
 - **Anti-Thundering Herd** — Mutex + protecție semafor împotriva furtunilor concurente de reîncercare
 - **Combo Fallback Chains** — Dacă furnizorul principal eșuează, trece automat prin lanț fără nicio intervenție
 - **Combo Circuit Breaker** — Dezactivează automat furnizorii care eșuează dintr-un lanț combinat
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Tabloul de bord pentru sănătate** — Monitorizare timp de funcționare, stări întrerupătoare de circuit, blocări, statistici cache, latență p50/p95/p99

 </details>
@@ -864,6 +864,7 @@ npm run electron:build:linux   # Linux (.AppImage)
 | Функция                          | Что делает                                                                   |
 | -------------------------------- | ---------------------------------------------------------------------------- |
 | 🔌 **Circuit Breaker**           | Авто-открытие/закрытие по провайдеру с настраиваемыми порогами               |
+| 🎯 **Endpoint-Aware Models**     | Custom models declare supported endpoints + API format                       |
 | 🛡️ **Anti-Thundering Herd**      | Mutex + семафор для API key провайдеров                                      |
 | 🧠 **Семантический кеш**         | Двухуровневый кеш (сигнатура + семантика) снижает стоимость                  |
 | ⚡ **Идемпотентность запросов**  | 5с окно дедупликации для дублирующихся запросов                              |
@@ -295,6 +295,7 @@ Poskytovatelia AI sa môžu stať nestabilnými, vrátiť chyby 5xx alebo dosiah
 - **Anti-Thundering Herd** - ochrana Mutex + semafor proti súbežným opakovaným búrkam
 - **Combo Fallback Chains** – Ak primárny poskytovateľ zlyhá, automaticky prepadne reťazcom bez akéhokoľvek zásahu
 - **Combo Circuit Breaker** – Automaticky deaktivuje zlyhávajúcich poskytovateľov v rámci kombinovaného reťazca
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
  – **Health Dashboard** – Monitorovanie dostupnosti, stavy ističov, blokovania, štatistiky vyrovnávacej pamäte, latencia p50/p95/p99

 </details>
@@ -864,6 +864,7 @@ npm run electron:build:linux   # Linux (.AppImage)
 | Funktion                               | Vad det gör                                                                       |
 | -------------------------------------- | --------------------------------------------------------------------------------- |
 | 🔌 **Circuit Breaker**                 | Autoöppna/stäng per leverantör med konfigurerbara trösklar                        |
+| 🎯 **Endpoint-Aware Models**           | Custom models declare supported endpoints + API format                            |
 | 🛡️ **Anti-ånflock**                    | Mutex + semaforhastighetsgräns för API-nyckelleverantörer                         |
 | 🧠 **Semantisk cache**                 | Tvåskiktscache (signatur + semantisk) minskar kostnaden och fördröjningen         |
 | ⚡ **Begär idempotens**                | 5s dedup-fönster för dubblettförfrågningar                                        |
@@ -294,6 +294,7 @@ OpenAI ใช้รูปแบบหนึ่ง Claude (Anthropic) ใช้
 - **Anti-Thundering Herd** — Mutex + การป้องกันเซมาฟอร์จากพายุที่ลองใหม่พร้อมกัน
 - **Combo Fallback Chains** — หากผู้ให้บริการหลักล้มเหลว จะตกผ่านห่วงโซ่โดยอัตโนมัติโดยไม่มีการแทรกแซง
 - **Combo Circuit Breaker** — ปิดการใช้งานผู้ให้บริการที่ล้มเหลวภายในคอมโบเชนโดยอัตโนมัติ
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **แดชบอร์ดสุขภาพ** — การตรวจสอบสถานะการออนไลน์ สถานะของเซอร์กิตเบรกเกอร์ การล็อก สถิติแคช เวลาแฝง p50/p95/p99

 </details>
@@ -294,6 +294,7 @@ OpenAI використовує один формат, Claude (Anthropic) вик
 - **Anti-Thundering Herd** — Mutex + захист семафора від одночасних повторних штормів
 - **Комбіновані запасні ланцюги** — якщо основний постачальник виходить з ладу, автоматично проходить через ланцюжок без втручання.
 - **Combo Circuit Breaker** — автоматично вимикає несправні постачальники в комбінованому ланцюжку
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Health Dashboard** — Моніторинг безвідмовної роботи, стани автоматичного вимикача, блокування, статистика кешу, затримка p50/p95/p99

 </details>
@@ -294,6 +294,7 @@ Các nhà cung cấp AI có thể trở nên không ổn định, trả về l
 - **Bầy chống sấm sét** — Mutex + bảo vệ semaphore chống lại các cơn bão thử lại đồng thời
 - **Chuỗi dự phòng kết hợp** — Nếu nhà cung cấp chính không thành công, nó sẽ tự động rơi qua chuỗi mà không cần can thiệp
 - **Combo Circuit Breaker** — Tự động vô hiệu hóa các nhà cung cấp bị lỗi trong chuỗi kết hợp
+  | 🎯 **Endpoint-Aware Models** | Custom models declare supported endpoints + API format |
 - **Bảng thông tin sức khỏe** — Giám sát thời gian hoạt động, trạng thái ngắt mạch, khóa, số liệu thống kê bộ nhớ đệm, độ trễ p50/p95/p99

 </details>
@@ -1348,15 +1348,16 @@ codex "your prompt"

 ## 📖 文档

-| 文档                                | 描述                         |
-| ----------------------------------- | ---------------------------- |
-| [用户指南](docs/USER_GUIDE.md)      | 提供商、组合、CLI 集成、部署 |
-| [API 参考](docs/API_REFERENCE.md)   | 所有端点及示例               |
-| [故障排除](docs/TROUBLESHOOTING.md) | 常见问题和解决方案           |
-| [架构](docs/ARCHITECTURE.md)        | 系统架构和内部机制           |
-| [贡献指南](CONTRIBUTING.md)         | 开发设置和指南               |
-| [OpenAPI 规范](docs/openapi.yaml)   | OpenAPI 3.0 规范             |
-| [安全策略](SECURITY.md)             | 漏洞报告和安全实践           |
+| 文档                                | 描述                                                   |
+| ----------------------------------- | ------------------------------------------------------ |
+| 🎯 **Endpoint-Aware Models**        | Custom models declare supported endpoints + API format |
+| [用户指南](docs/USER_GUIDE.md)      | 提供商、组合、CLI 集成、部署                           |
+| [API 参考](docs/API_REFERENCE.md)   | 所有端点及示例                                         |
+| [故障排除](docs/TROUBLESHOOTING.md) | 常见问题和解决方案                                     |
+| [架构](docs/ARCHITECTURE.md)        | 系统架构和内部机制                                     |
+| [贡献指南](CONTRIBUTING.md)         | 开发设置和指南                                         |
+| [OpenAPI 规范](docs/openapi.yaml)   | OpenAPI 3.0 规范                                       |
+| [安全策略](SECURITY.md)             | 漏洞报告和安全实践                                     |

 ---

@@ -26,10 +26,12 @@ const {
  nativeImage,
  shell,
  session,
+  Notification,
 } = require("electron");
 const path = require("path");
 const { spawn } = require("child_process");
 const fs = require("fs");
+const { autoUpdater } = require("electron-updater");

 // ── Single Instance Lock ───────────────────────────────────
 const gotTheLock = app.requestSingleInstanceLock();
@@ -62,6 +64,11 @@ let serverPort = 20128;

 const getServerUrl = () => `http://localhost:${serverPort}`;

+// ── Auto-Updater Configuration ──────────────────────────────
+autoUpdater.autoDownload = false;
+autoUpdater.autoInstallOnAppQuit = true;
+autoUpdater.logger = console;
+
 // ── Helper: Send IPC event to renderer (#5) ────────────────
 function sendToRenderer(channel, data) {
  if (mainWindow && !mainWindow.isDestroyed()) {
@@ -103,6 +110,77 @@ async function waitForServerExit(proc, timeoutMs = 5000) {
  ]);
 }

+// ── Auto-Updater Event Handlers ─────────────────────────────
+function setupAutoUpdater() {
+  autoUpdater.on("checking-for-update", () => {
+    sendToRenderer("update-status", { status: "checking" });
+    console.log("[Electron] Checking for updates...");
+  });
+
+  autoUpdater.on("update-available", (info) => {
+    sendToRenderer("update-status", { status: "available", version: info.version });
+    console.log("[Electron] Update available:", info.version);
+  });
+
+  autoUpdater.on("update-not-available", (info) => {
+    sendToRenderer("update-status", { status: "not-available", version: info.version });
+    console.log("[Electron] No update available");
+  });
+
+  autoUpdater.on("download-progress", (progress) => {
+    sendToRenderer("update-status", {
+      status: "downloading",
+      percent: Math.round(progress.percent),
+      transferred: progress.transferred,
+      total: progress.total,
+    });
+  });
+
+  autoUpdater.on("update-downloaded", (info) => {
+    sendToRenderer("update-status", { status: "downloaded", version: info.version });
+    console.log("[Electron] Update downloaded:", info.version);
+
+    if (Notification.isSupported()) {
+      const notification = new Notification({
+        title: "OmniRoute Update Ready",
+        body: `Version ${info.version} is ready to install. Click to restart.`,
+      });
+      notification.on("click", () => {
+        autoUpdater.quitAndInstall();
+      });
+      notification.show();
+    }
+  });
+
+  autoUpdater.on("error", (error) => {
+    sendToRenderer("update-status", { status: "error", message: error.message });
+    console.error("[Electron] Update error:", error);
+  });
+}
+
+async function checkForUpdates(silent = false) {
+  if (isDev) {
+    console.log("[Electron] Dev mode — skipping auto-update");
+    if (!silent) {
+      sendToRenderer("update-status", { status: "error", message: "Updates disabled in dev mode" });
+    }
+    return;
+  }
+  await autoUpdater.checkForUpdates();
+}
+
+async function downloadUpdate() {
+  await autoUpdater.downloadUpdate();
+}
+
+function installUpdate() {
+  if (nextServer) {
+    nextServer.kill("SIGTERM");
+    nextServer = null;
+  }
+  autoUpdater.quitAndInstall();
+}
+
 // ── Content Security Policy (#15) ──────────────────────────
 function setupContentSecurityPolicy() {
  session.defaultSession.webRequest.onHeadersReceived((details, callback) => {
@@ -236,6 +314,11 @@ function createTray() {
      ],
    },
    { type: "separator" },
+    {
+      label: "Check for Updates",
+      click: () => checkForUpdates(false),
+    },
+    { type: "separator" },
    {
      label: "Quit",
      click: () => {
@@ -391,6 +474,36 @@ function setupIpcHandlers() {
  });

  ipcMain.on("window-close", () => mainWindow?.close());
+
+  // Auto-update IPC handlers
+  ipcMain.handle("check-for-updates", async () => {
+    try {
+      await checkForUpdates(false);
+      return { success: true };
+    } catch (error) {
+      console.error("[Electron] Check for updates failed:", error);
+      sendToRenderer("update-status", { status: "error", message: error.message });
+      return { success: false, error: error.message };
+    }
+  });
+
+  ipcMain.handle("download-update", async () => {
+    try {
+      await downloadUpdate();
+      return { success: true };
+    } catch (error) {
+      console.error("[Electron] Download update failed:", error);
+      sendToRenderer("update-status", { status: "error", message: error.message });
+      return { success: false, error: error.message };
+    }
+  });
+
+  ipcMain.handle("install-update", () => {
+    installUpdate();
+    // No return value — app will quit and restart
+  });
+
+  ipcMain.handle("get-app-version", () => app.getVersion());
 }

 // ── App Lifecycle ──────────────────────────────────────────
@@ -407,6 +520,14 @@ app.whenReady().then(async () => {
  createWindow();
  createTray();
  setupIpcHandlers();
+  setupAutoUpdater();
+
+  // Check for updates after a short delay (don't block startup)
+  if (!isDev) {
+    setTimeout(() => {
+      checkForUpdates(true);
+    }, 3000);
+  }

  // macOS: recreate window when dock icon clicked
  app.on("activate", () => {
@@ -15,7 +15,9 @@
    "build:linux": "electron-builder --linux",
    "pack": "electron-builder --dir"
  },
-  "dependencies": {},
+  "dependencies": {
+    "electron-updater": "^6.8.3"
+  },
  "devDependencies": {
    "electron": "^40.6.1",
    "electron-builder": "^25.1.8"
@@ -28,6 +30,11 @@
      "output": "dist-electron",
      "buildResources": "assets"
    },
+    "publish": {
+      "provider": "github",
+      "owner": "diegosouzapw",
+      "repo": "OmniRoute"
+    },
    "files": [
      "main.js",
      "preload.js",
@@ -13,9 +13,18 @@ const { contextBridge, ipcRenderer } = require("electron");

 // ── Channel Whitelist ──────────────────────────────────────
 const VALID_CHANNELS = {
-  invoke: ["get-app-info", "open-external", "get-data-dir", "restart-server"],
+  invoke: [
+    "get-app-info",
+    "open-external",
+    "get-data-dir",
+    "restart-server",
+    "check-for-updates",
+    "download-update",
+    "install-update",
+    "get-app-version",
+  ],
  send: ["window-minimize", "window-maximize", "window-close"],
-  receive: ["server-status", "port-changed"],
+  receive: ["server-status", "port-changed", "update-status"],
 };

 // ── Fix #16: Generic IPC wrappers ──────────────────────────
@@ -48,6 +57,12 @@ contextBridge.exposeInMainWorld("electronAPI", {
  openExternal: (url) => safeInvoke("open-external", url),
  getDataDir: () => safeInvoke("get-data-dir"),
  restartServer: () => safeInvoke("restart-server"),
+  getAppVersion: () => safeInvoke("get-app-version"),
+
+  // ── Auto-Update ──────────────────────────────────────────
+  checkForUpdates: () => safeInvoke("check-for-updates"),
+  downloadUpdate: () => safeInvoke("download-update"),
+  installUpdate: () => safeInvoke("install-update"),

  // ── Send (fire-and-forget) ───────────────────────────────
  minimizeWindow: () => safeSend("window-minimize"),
@@ -58,6 +73,7 @@ contextBridge.exposeInMainWorld("electronAPI", {
  // Fix #6: Returns a disposer function for precise cleanup
  onServerStatus: (callback) => safeOn("server-status", callback),
  onPortChanged: (callback) => safeOn("port-changed", callback),
+  onUpdateStatus: (callback) => safeOn("update-status", callback),

  // ── Static Properties ────────────────────────────────────
  isElectron: true,
@@ -6,7 +6,7 @@ const withNextIntl = createNextIntlPlugin("./src/i18n/request.ts");
 const nextConfig = {
  turbopack: {},
  output: "standalone",
-  serverExternalPackages: ["better-sqlite3"],
+  serverExternalPackages: ["better-sqlite3", "zod"],
  transpilePackages: ["@omniroute/open-sse"],
  allowedDevOrigins: ["192.168.*"],
  typescript: {
@@ -0,0 +1,264 @@
+/**
+ * CLI Fingerprint Definitions
+ *
+ * Defines per-provider "fingerprints" that control the exact ordering of HTTP headers
+ * and JSON body fields to match the native CLI tools exactly.
+ *
+ * When `cliCompatMode` is enabled for a provider, OmniRoute reorders outgoing requests
+ * to be indistinguishable from the real CLI binary, reducing account flagging risk.
+ *
+ * Header order and body field order were captured via mitmproxy traffic analysis.
+ */
+
+export interface CliFingerprint {
+  /** Ordered list of header names (case-sensitive). Unlisted headers are appended. */
+  headerOrder: string[];
+  /** Ordered list of top-level JSON body fields. Unlisted fields are appended. */
+  bodyFieldOrder: string[];
+  /** User-Agent string to inject (overrides default) */
+  userAgent?: string;
+  /** Extra headers to add */
+  extraHeaders?: Record<string, string>;
+}
+
+/**
+ * Fingerprint registry - keyed by provider alias (lowercase).
+ * Based on mitmproxy traffic captures from native CLI tools.
+ */
+export const CLI_FINGERPRINTS: Record<string, CliFingerprint> = {
+  codex: {
+    headerOrder: [
+      "Host",
+      "Content-Type",
+      "Authorization",
+      "Accept",
+      "User-Agent",
+      "Accept-Encoding",
+    ],
+    bodyFieldOrder: [
+      "model",
+      "messages",
+      "temperature",
+      "top_p",
+      "max_tokens",
+      "stream",
+      "tools",
+      "tool_choice",
+      "response_format",
+      "n",
+      "stop",
+    ],
+    userAgent: "codex-cli",
+  },
+  claude: {
+    headerOrder: [
+      "Host",
+      "Content-Type",
+      "x-api-key",
+      "anthropic-version",
+      "Accept",
+      "User-Agent",
+      "Accept-Encoding",
+    ],
+    bodyFieldOrder: [
+      "model",
+      "max_tokens",
+      "messages",
+      "system",
+      "temperature",
+      "top_p",
+      "top_k",
+      "stream",
+      "tools",
+      "tool_choice",
+      "metadata",
+    ],
+    userAgent: "claude-code",
+  },
+  github: {
+    headerOrder: [
+      "Host",
+      "Authorization",
+      "X-Request-Id",
+      "Vscode-Sessionid",
+      "Vscode-Machineid",
+      "Editor-Version",
+      "Editor-Plugin-Version",
+      "Copilot-Integration-Id",
+      "Openai-Organization",
+      "Openai-Intent",
+      "Content-Type",
+      "User-Agent",
+      "Accept",
+      "Accept-Encoding",
+    ],
+    bodyFieldOrder: [
+      "messages",
+      "model",
+      "temperature",
+      "top_p",
+      "max_tokens",
+      "n",
+      "stream",
+      "intent",
+      "intent_threshold",
+      "intent_content",
+    ],
+    userAgent: "GitHubCopilotChat",
+  },
+  antigravity: {
+    headerOrder: [
+      "Host",
+      "Content-Type",
+      "Authorization",
+      "User-Agent",
+      "Accept",
+      "Accept-Encoding",
+    ],
+    bodyFieldOrder: ["project", "model", "userAgent", "requestType", "requestId", "request"],
+    userAgent: "antigravity",
+  },
+};
+
+/**
+ * Reorder an object's keys according to a specified order.
+ * Keys not in the order list are appended at the end in their original order.
+ */
+export function orderFields<T extends Record<string, unknown>>(obj: T, fieldOrder: string[]): T {
+  if (!fieldOrder?.length || !obj || typeof obj !== "object") return obj;
+
+  const result: Record<string, unknown> = {};
+  const remaining = new Set(Object.keys(obj));
+
+  // First, add fields in the specified order
+  for (const key of fieldOrder) {
+    if (key in obj) {
+      result[key] = obj[key];
+      remaining.delete(key);
+    }
+  }
+
+  // Then append remaining fields in original order
+  for (const key of remaining) {
+    result[key] = obj[key];
+  }
+
+  return result as T;
+}
+
+/**
+ * Reorder HTTP headers according to a fingerprint.
+ * Returns a new object with headers in the specified order.
+ */
+export function orderHeaders(
+  headers: Record<string, string>,
+  headerOrder: string[]
+): Record<string, string> {
+  if (!headerOrder?.length || !headers) return headers;
+
+  const result: Record<string, string> = {};
+  const remaining = new Map<string, string>();
+
+  // Build case-insensitive lookup
+  const headerMap = new Map<string, [string, string]>();
+  for (const [key, value] of Object.entries(headers)) {
+    headerMap.set(key.toLowerCase(), [key, value]);
+  }
+
+  // Add ordered headers first
+  for (const orderedKey of headerOrder) {
+    const entry = headerMap.get(orderedKey.toLowerCase());
+    if (entry) {
+      result[entry[0]] = entry[1];
+      headerMap.delete(orderedKey.toLowerCase());
+    }
+  }
+
+  // Add remaining headers
+  for (const [, [key, value]] of headerMap) {
+    result[key] = value;
+  }
+
+  return result;
+}
+
+/**
+ * Apply a CLI fingerprint to headers and body.
+ * Returns { headers, bodyString } with the correct ordering.
+ */
+export function applyFingerprint(
+  provider: string,
+  headers: Record<string, string>,
+  body: unknown
+): { headers: Record<string, string>; bodyString: string } {
+  const fingerprint = CLI_FINGERPRINTS[provider?.toLowerCase()];
+
+  if (!fingerprint) {
+    return { headers, bodyString: JSON.stringify(body) };
+  }
+
+  // Apply user agent override
+  if (fingerprint.userAgent) {
+    headers["User-Agent"] = fingerprint.userAgent;
+  }
+
+  // Apply extra headers
+  if (fingerprint.extraHeaders) {
+    Object.assign(headers, fingerprint.extraHeaders);
+  }
+
+  // Reorder headers
+  const orderedHeaders = orderHeaders(headers, fingerprint.headerOrder);
+
+  // Reorder body fields
+  const orderedBody =
+    body && typeof body === "object" && !Array.isArray(body)
+      ? orderFields(body as Record<string, unknown>, fingerprint.bodyFieldOrder)
+      : body;
+
+  return {
+    headers: orderedHeaders,
+    bodyString: JSON.stringify(orderedBody),
+  };
+}
+
+/**
+ * Runtime cache for CLI compat providers set via Settings UI.
+ * Updated by the settings API when users toggle providers.
+ */
+let _cliCompatProviders: Set<string> = new Set();
+
+/**
+ * Update the runtime cache of CLI-compat-enabled providers.
+ * Called from the settings API when cliCompatProviders is updated.
+ */
+export function setCliCompatProviders(providers: string[]): void {
+  _cliCompatProviders = new Set((providers || []).map((p) => p.toLowerCase()));
+}
+
+/**
+ * Get the current list of CLI-compat-enabled providers.
+ */
+export function getCliCompatProviders(): string[] {
+  return Array.from(_cliCompatProviders);
+}
+
+/**
+ * Check if CLI compatibility mode is enabled for a provider.
+ * Reads from: 1) Runtime cache (Settings UI), 2) Environment variables.
+ */
+export function isCliCompatEnabled(provider: string): boolean {
+  const key = provider?.toLowerCase().replace(/[^a-z0-9]/g, "_");
+
+  // 1. Check runtime cache (set via Settings UI)
+  if (_cliCompatProviders.has(provider?.toLowerCase())) return true;
+
+  // 2. Check environment variable: CLI_COMPAT_<PROVIDER>=1
+  const envKey = `CLI_COMPAT_${key?.toUpperCase()}`;
+  if (process.env[envKey] === "1" || process.env[envKey] === "true") return true;
+
+  // 3. Global enable: CLI_COMPAT_ALL=1
+  if (process.env.CLI_COMPAT_ALL === "1" || process.env.CLI_COMPAT_ALL === "true") return true;
+
+  return false;
+}
@@ -1,4 +1,5 @@
 import { HTTP_STATUS, FETCH_TIMEOUT_MS } from "../config/constants.ts";
+import { applyFingerprint, isCliCompatEnabled } from "../config/cliFingerprints.ts";

 type JsonRecord = Record<string, unknown>;

@@ -192,10 +193,20 @@ export class BaseExecutor {
            ? mergeAbortSignals(signal, timeoutSignal)
            : signal || timeoutSignal;

+        // Apply CLI fingerprint ordering if enabled for this provider
+        let finalHeaders = headers;
+        let bodyString = JSON.stringify(transformedBody);
+
+        if (isCliCompatEnabled(this.provider)) {
+          const fingerprinted = applyFingerprint(this.provider, headers, transformedBody);
+          finalHeaders = fingerprinted.headers;
+          bodyString = fingerprinted.bodyString;
+        }
+
        const fetchOptions: RequestInit = {
          method: "POST",
-          headers,
-          body: JSON.stringify(transformedBody),
+          headers: finalHeaders,
+          body: bodyString,
        };
        if (combinedSignal) fetchOptions.signal = combinedSignal;

@@ -30,9 +30,23 @@ import {
 * @param {object} options.body - Request body
 * @param {object} options.credentials - Provider credentials { apiKey, accessToken }
 * @param {object} options.log - Logger
+ * @param {string} [options.resolvedProvider] - Pre-resolved provider ID (from route layer custom model resolution)
 */
-export async function handleImageGeneration({ body, credentials, log }) {
-  const { provider, model } = parseImageModel(body.model);
+export async function handleImageGeneration({ body, credentials, log, resolvedProvider = null }) {
+  let provider, model;
+
+  if (resolvedProvider) {
+    // Provider was already resolved by the route layer (custom model from DB)
+    // Extract model name from the full "provider/model" string
+    provider = resolvedProvider;
+    const modelStr = body.model || "";
+    model = modelStr.startsWith(provider + "/") ? modelStr.slice(provider.length + 1) : modelStr;
+  } else {
+    // Standard path: resolve from built-in image registry
+    const parsed = parseImageModel(body.model);
+    provider = parsed.provider;
+    model = parsed.model;
+  }

  if (!provider) {
    return {
@@ -43,12 +57,42 @@ export async function handleImageGeneration({ body, credentials, log }) {
  }

  const providerConfig = getImageProvider(provider);
+
+  // For custom models without a built-in provider config, use OpenAI-compatible handler
+  // with a synthetic config based on the provider's credentials
  if (!providerConfig) {
-    return {
-      success: false,
-      status: 400,
-      error: `Unknown image provider: ${provider}`,
+    if (!resolvedProvider) {
+      return {
+        success: false,
+        status: 400,
+        error: `Unknown image provider: ${provider}`,
+      };
+    }
+
+    // Custom model: use OpenAI-compatible format with provider's base URL
+    // The credentials were already resolved by the route layer
+    if (log) {
+      log.info("IMAGE", `Custom model ${provider}/${model} — using OpenAI-compatible handler`);
+    }
+
+    const syntheticConfig = {
+      id: provider,
+      baseUrl:
+        credentials?.baseUrl ||
+        `https://generativelanguage.googleapis.com/v1beta/openai/images/generations`,
+      authType: "apikey",
+      authHeader: "bearer",
+      format: "openai",
    };
+
+    return handleOpenAIImageGeneration({
+      model,
+      provider,
+      providerConfig: syntheticConfig,
+      body,
+      credentials,
+      log,
+    });
  }

  // Route to format-specific handler
@@ -310,7 +310,8 @@ export async function handleComboChat({
    const parsed = parseModel(modelStr);
    const provider = parsed.provider || parsed.providerAlias || "unknown";
    const profile = getProviderProfile(provider);
-    const breaker = getCircuitBreaker(`combo:${provider}`, {
+    const breakerKey = `combo:${modelStr}`;
+    const breaker = getCircuitBreaker(breakerKey, {
      failureThreshold: profile.circuitBreakerThreshold,
      resetTimeout: profile.circuitBreakerReset,
    });
@@ -440,8 +441,7 @@ export async function handleComboChat({

  // Early exit: check if all models have breaker OPEN
  const allBreakersOpen = orderedModels.every((m) => {
-    const p = parseModel(m).provider || parseModel(m).providerAlias || "unknown";
-    return !getCircuitBreaker(`combo:${p}`).canExecute();
+    return !getCircuitBreaker(`combo:${m}`).canExecute();
  });

  // All models failed
@@ -532,7 +532,8 @@ async function handleRoundRobinCombo({
    const parsed = parseModel(modelStr);
    const provider = parsed.provider || parsed.providerAlias || "unknown";
    const profile = getProviderProfile(provider);
-    const breaker = getCircuitBreaker(`combo:${provider}`, {
+    const breakerKey = `combo:${modelStr}`;
+    const breaker = getCircuitBreaker(breakerKey, {
      failureThreshold: profile.circuitBreakerThreshold,
      resetTimeout: profile.circuitBreakerReset,
    });
@@ -694,8 +695,7 @@ async function handleRoundRobinCombo({

  // Early exit: check if all models have breaker OPEN
  const allBreakersOpen = orderedModels.every((m) => {
-    const p = parseModel(m).provider || parseModel(m).providerAlias || "unknown";
-    return !getCircuitBreaker(`combo:${p}`).canExecute();
+    return !getCircuitBreaker(`combo:${m}`).canExecute();
  });

  if (allBreakersOpen) {
@@ -3,6 +3,7 @@
 */

 import { PROVIDERS } from "../config/constants.ts";
+import { safePercentage } from "@/shared/utils/formatting";

 // GitHub API config
 const GITHUB_CONFIG = {
@@ -34,6 +35,7 @@ const CLAUDE_CONFIG = {
  oauthUsageUrl: "https://api.anthropic.com/api/oauth/usage",
  usageUrl: "https://api.anthropic.com/v1/organizations/{org_id}/usage",
  settingsUrl: "https://api.anthropic.com/v1/settings",
+  apiVersion: "2023-06-01",
 };

 type JsonRecord = Record<string, unknown>;
@@ -469,7 +471,7 @@ async function getClaudeUsage(accessToken) {
      headers: {
        Authorization: `Bearer ${accessToken}`,
        "anthropic-beta": "oauth-2025-04-20",
-        "anthropic-version": "2023-06-01",
+        "anthropic-version": CLAUDE_CONFIG.apiVersion,
      },
    });

@@ -477,36 +479,34 @@ async function getClaudeUsage(accessToken) {
      const data = await oauthResponse.json();
      const quotas: Record<string, any> = {};

-      // utilization = percentage USED (e.g., 22 means 22% used, 78% remaining)
+      // utilization = percentage REMAINING (e.g., 90 means 90% remaining, 10% used)
+      const hasUtilization = (window: any) =>
+        window && typeof window === "object" && safePercentage(window.utilization) !== undefined;
+
      const createQuotaObject = (window: any) => {
-        const used = window?.utilization ?? 0;
-        const remaining = 100 - used;
+        const remaining = safePercentage(window.utilization) as number;
+        const used = 100 - remaining;
        return {
          used,
          total: 100,
          remaining,
-          resetAt: parseResetTime(window?.resets_at),
+          resetAt: parseResetTime(window.resets_at),
          remainingPercentage: remaining,
          unlimited: false,
        };
      };

-      if (data.five_hour && typeof data.five_hour === "object") {
+      if (hasUtilization(data.five_hour)) {
        quotas["session (5h)"] = createQuotaObject(data.five_hour);
      }

-      if (data.seven_day && typeof data.seven_day === "object") {
+      if (hasUtilization(data.seven_day)) {
        quotas["weekly (7d)"] = createQuotaObject(data.seven_day);
      }

      // Parse model-specific weekly windows (e.g., seven_day_sonnet, seven_day_opus)
      for (const [key, value] of Object.entries(data)) {
-        if (
-          key.startsWith("seven_day_") &&
-          key !== "seven_day" &&
-          value &&
-          typeof value === "object"
-        ) {
+        if (key.startsWith("seven_day_") && key !== "seven_day" && hasUtilization(value)) {
          const modelName = key.replace("seven_day_", "");
          quotas[`weekly ${modelName} (7d)`] = createQuotaObject(value);
        }
@@ -519,7 +519,10 @@ async function getClaudeUsage(accessToken) {
      };
    }

-    // Fallback: Try legacy settings/org endpoint (for API key users with org admin access)
+    // Fallback: OAuth endpoint returned non-OK, try legacy settings/org endpoint
+    console.warn(
+      `[Claude Usage] OAuth endpoint returned ${oauthResponse.status}, falling back to legacy`
+    );
    return await getClaudeUsageLegacy(accessToken);
  } catch (error) {
    return { message: `Claude connected. Unable to fetch usage: ${(error as any).message}` };
@@ -536,7 +539,7 @@ async function getClaudeUsageLegacy(accessToken) {
      method: "GET",
      headers: {
        Authorization: `Bearer ${accessToken}`,
-        "anthropic-version": "2023-06-01",
+        "anthropic-version": CLAUDE_CONFIG.apiVersion,
      },
    });

@@ -550,7 +553,7 @@ async function getClaudeUsageLegacy(accessToken) {
            method: "GET",
            headers: {
              Authorization: `Bearer ${accessToken}`,
-              "anthropic-version": "2023-06-01",
+              "anthropic-version": CLAUDE_CONFIG.apiVersion,
            },
          }
        );
@@ -185,15 +185,19 @@ export function prepareClaudeRequest(body, provider = null) {
    }
  }

-  // 3. Tools: remove all cache_control, add only to last tool with ttl 1h
+  // 3. Tools: remove all cache_control, add only to last non-deferred tool with ttl 1h
+  // Tools with defer_loading=true cannot have cache_control (API rejects it)
  if (body.tools && Array.isArray(body.tools)) {
-    body.tools = body.tools.map((tool, i) => {
+    body.tools = body.tools.map((tool) => {
      const { cache_control, ...rest } = tool;
-      if (i === body.tools.length - 1) {
-        return { ...rest, cache_control: { type: "ephemeral", ttl: "1h" } };
-      }
      return rest;
    });
+    for (let i = body.tools.length - 1; i >= 0; i--) {
+      if (!body.tools[i].defer_loading) {
+        body.tools[i].cache_control = { type: "ephemeral", ttl: "1h" };
+        break;
+      }
+    }
  }

  return body;
@@ -275,30 +275,11 @@ export function openaiToOpenAIResponsesRequest(

    // Convert assistant messages
    if (role === "assistant") {
-      // Add reasoning content before assistant output
-      if (msg.reasoning_content) {
-        input.push({
-          type: "reasoning",
-          id: `reasoning_${input.length}`,
-          summary: [{ type: "summary_text", text: toString(msg.reasoning_content) }],
-        });
-      }
+      // Skip reasoning_content — OpenAI Responses API requires server-generated
+      // rs_* IDs for reasoning items. Synthesizing client-side IDs (e.g. reasoning_N)
+      // causes 400 errors from Responses-compatible upstreams. (#224)

-      // Handle thinking blocks in array content
-      if (Array.isArray(msg.content)) {
-        for (const blockValue of msg.content) {
-          const block = toRecord(blockValue);
-          if (block.type === "thinking" || block.type === "redacted_thinking") {
-            input.push({
-              type: "reasoning",
-              id: `reasoning_${input.length}`,
-              summary: [
-                { type: "summary_text", text: toString(block.thinking || block.data, "...") },
-              ],
-            });
-          }
-        }
-      }
+      // Skip thinking blocks in array content — same rs_* ID constraint applies

      // Build assistant output content
      const outputContent: unknown[] = [];
@@ -175,8 +175,13 @@ export function openaiToClaudeRequest(model, body, stream) {
      };
    });

-    if (result.tools.length > 0) {
-      result.tools[result.tools.length - 1].cache_control = { type: "ephemeral", ttl: "1h" };
+    // Add cache_control to last tool that doesn't have defer_loading
+    // Tools with defer_loading=true cannot have cache_control (API rejects it)
+    for (let i = result.tools.length - 1; i >= 0; i--) {
+      if (!result.tools[i].defer_loading) {
+        result.tools[i].cache_control = { type: "ephemeral", ttl: "1h" };
+        break;
+      }
    }
  }

@@ -1,12 +1,12 @@
 {
  "name": "omniroute",
-  "version": "2.0.0",
+  "version": "2.0.9",
  "lockfileVersion": 3,
  "requires": true,
  "packages": {
    "": {
      "name": "omniroute",
-      "version": "2.0.0",
+      "version": "2.0.9",
      "hasInstallScript": true,
      "license": "MIT",
      "workspaces": [
@@ -6596,12 +6596,12 @@
      }
    },
    "node_modules/express-rate-limit": {
-      "version": "8.2.1",
-      "resolved": "https://registry.npmjs.org/express-rate-limit/-/express-rate-limit-8.2.1.tgz",
-      "integrity": "sha512-PCZEIEIxqwhzw4KF0n7QF4QqruVTcF73O5kFKUnGOyjbCCgizBBiFaYpd/fnBLUMPw/BWw9OsiN7GgrNYr7j6g==",
+      "version": "8.3.0",
+      "resolved": "https://registry.npmjs.org/express-rate-limit/-/express-rate-limit-8.3.0.tgz",
+      "integrity": "sha512-KJzBawY6fB9FiZGdE/0aftepZ91YlaGIrV8vgblRM3J8X+dHx/aiowJWwkx6LIGyuqGiANsjSwwrbb8mifOJ4Q==",
      "license": "MIT",
      "dependencies": {
-        "ip-address": "10.0.1"
+        "ip-address": "10.1.0"
      },
      "engines": {
        "node": ">= 16"
@@ -6613,15 +6613,6 @@
        "express": ">= 4.11"
      }
    },
-    "node_modules/express-rate-limit/node_modules/ip-address": {
-      "version": "10.0.1",
-      "resolved": "https://registry.npmjs.org/ip-address/-/ip-address-10.0.1.tgz",
-      "integrity": "sha512-NWv9YLW4PoW2B7xtzaS3NCot75m6nK7Icdv0o3lfMceJVRfSoQwqD4wEH5rLwoKJwUiZ/rfpiVBhnaF0FK4HoA==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 12"
-      }
-    },
    "node_modules/fast-copy": {
      "version": "4.0.2",
      "resolved": "https://registry.npmjs.org/fast-copy/-/fast-copy-4.0.2.tgz",
@@ -1,6 +1,6 @@
 {
  "name": "omniroute",
-  "version": "2.0.0",
+  "version": "2.0.10",
  "description": "Smart AI Router with auto fallback — route to FREE & cheap models, zero downtime. Works with Cursor, Cline, Claude Desktop, Codex, and any OpenAI-compatible tool.",
  "type": "module",
  "bin": {
@@ -0,0 +1,400 @@
+"use client";
+
+import { useState, useEffect, useCallback, useRef } from "react";
+import { Card, Button, Select, Badge } from "@/shared/components";
+import dynamic from "next/dynamic";
+
+const Editor = dynamic(() => import("@monaco-editor/react"), { ssr: false });
+
+interface ModelInfo {
+  id: string;
+  object: string;
+  owned_by: string;
+}
+
+interface ProviderOption {
+  value: string;
+  label: string;
+}
+
+const ENDPOINT_OPTIONS = [
+  { value: "chat", label: "Chat Completions" },
+  { value: "responses", label: "Responses" },
+  { value: "images", label: "Image Generation" },
+  { value: "embeddings", label: "Embeddings" },
+];
+
+const DEFAULT_BODIES: Record<string, object> = {
+  chat: {
+    model: "",
+    messages: [{ role: "user", content: "Hello! Say hi in one sentence." }],
+    max_tokens: 100,
+    stream: false,
+  },
+  responses: {
+    model: "",
+    input: "Hello! Say hi in one sentence.",
+    stream: false,
+  },
+  images: {
+    model: "",
+    prompt: "A beautiful sunset over mountains",
+    n: 1,
+    size: "1024x1024",
+  },
+  embeddings: {
+    model: "",
+    input: "Hello world",
+    encoding_format: "float",
+  },
+};
+
+const ENDPOINT_PATHS: Record<string, string> = {
+  chat: "/v1/chat/completions",
+  responses: "/v1/responses",
+  images: "/v1/images/generations",
+  embeddings: "/v1/embeddings",
+};
+
+export default function PlaygroundPage() {
+  const [models, setModels] = useState<ModelInfo[]>([]);
+  const [providers, setProviders] = useState<ProviderOption[]>([]);
+  const [selectedProvider, setSelectedProvider] = useState("");
+  const [selectedModel, setSelectedModel] = useState("");
+  const [selectedEndpoint, setSelectedEndpoint] = useState("chat");
+  const [requestBody, setRequestBody] = useState("");
+  const [responseBody, setResponseBody] = useState("");
+  const [loading, setLoading] = useState(false);
+  const [responseStatus, setResponseStatus] = useState<number | null>(null);
+  const [responseDuration, setResponseDuration] = useState<number | null>(null);
+  const abortRef = useRef<AbortController | null>(null);
+
+  // Fetch models
+  useEffect(() => {
+    fetch("/v1/models")
+      .then((res) => res.json())
+      .then((data) => {
+        const modelList = (data?.data || []) as ModelInfo[];
+        setModels(modelList);
+
+        // Extract unique providers from model ids (provider/model format)
+        const providerSet = new Set<string>();
+        modelList.forEach((m) => {
+          const parts = m.id.split("/");
+          if (parts.length >= 2) providerSet.add(parts[0]);
+        });
+        const providerOpts = Array.from(providerSet)
+          .sort()
+          .map((p) => ({ value: p, label: p }));
+        setProviders(providerOpts);
+        if (providerOpts.length > 0) setSelectedProvider(providerOpts[0].value);
+      })
+      .catch(() => {});
+  }, []);
+
+  // Filter models by selected provider
+  const filteredModels = models
+    .filter((m) => !selectedProvider || m.id.startsWith(selectedProvider + "/"))
+    .map((m) => ({ value: m.id, label: m.id }));
+
+  // Helper to generate default body for a given endpoint and model
+  const generateDefaultBody = (endpoint: string, model: string) => {
+    const template = { ...DEFAULT_BODIES[endpoint] };
+    if ("model" in template) {
+      (template as any).model = model;
+    }
+    return JSON.stringify(template, null, 2);
+  };
+
+  // When provider changes, auto-select first model and reset body
+  const handleProviderChange = (newProvider: string) => {
+    setSelectedProvider(newProvider);
+    const providerModels = models
+      .filter((m) => !newProvider || m.id.startsWith(newProvider + "/"))
+      .map((m) => m.id);
+    const firstModel = providerModels[0] || "";
+    setSelectedModel(firstModel);
+    setRequestBody(generateDefaultBody(selectedEndpoint, firstModel));
+    setResponseBody("");
+    setResponseStatus(null);
+    setResponseDuration(null);
+  };
+
+  // When model changes, update body
+  const handleModelChange = (newModel: string) => {
+    setSelectedModel(newModel);
+    setRequestBody(generateDefaultBody(selectedEndpoint, newModel));
+    setResponseBody("");
+    setResponseStatus(null);
+    setResponseDuration(null);
+  };
+
+  // When endpoint changes, update body
+  const handleEndpointChange = (newEndpoint: string) => {
+    setSelectedEndpoint(newEndpoint);
+    setRequestBody(generateDefaultBody(newEndpoint, selectedModel));
+    setResponseBody("");
+    setResponseStatus(null);
+    setResponseDuration(null);
+  };
+
+  const handleSend = useCallback(async () => {
+    if (!requestBody.trim()) return;
+    setLoading(true);
+    setResponseBody("");
+    setResponseStatus(null);
+    setResponseDuration(null);
+
+    const controller = new AbortController();
+    abortRef.current = controller;
+    const startTime = Date.now();
+
+    try {
+      const parsed = JSON.parse(requestBody);
+      const path = ENDPOINT_PATHS[selectedEndpoint];
+      const res = await fetch(path, {
+        method: "POST",
+        headers: { "Content-Type": "application/json" },
+        body: JSON.stringify(parsed),
+        signal: controller.signal,
+      });
+
+      setResponseStatus(res.status);
+      setResponseDuration(Date.now() - startTime);
+
+      const contentType = res.headers.get("content-type") || "";
+      if (contentType.includes("text/event-stream")) {
+        // Handle streaming
+        const reader = res.body?.getReader();
+        const decoder = new TextDecoder();
+        let accumulated = "";
+        if (reader) {
+          while (true) {
+            const { done, value } = await reader.read();
+            if (done) break;
+            accumulated += decoder.decode(value, { stream: true });
+            setResponseBody(accumulated);
+          }
+        }
+      } else {
+        const data = await res.json();
+        setResponseBody(JSON.stringify(data, null, 2));
+      }
+    } catch (err: any) {
+      if (err.name === "AbortError") {
+        setResponseBody(JSON.stringify({ cancelled: true }, null, 2));
+      } else {
+        setResponseBody(JSON.stringify({ error: err.message }, null, 2));
+      }
+      setResponseDuration(Date.now() - startTime);
+    }
+    setLoading(false);
+  }, [requestBody, selectedEndpoint]);
+
+  const handleCancel = () => {
+    if (abortRef.current) {
+      abortRef.current.abort();
+    }
+  };
+
+  const handleCopy = async (text: string) => {
+    try {
+      await navigator.clipboard.writeText(text);
+    } catch {
+      /* silent */
+    }
+  };
+
+  return (
+    <div className="space-y-5">
+      {/* Info Banner */}
+      <div className="flex items-start gap-3 px-4 py-3 rounded-lg bg-primary/5 border border-primary/10 text-sm text-text-muted">
+        <span className="material-symbols-outlined text-primary text-[20px] mt-0.5 shrink-0">
+          science
+        </span>
+        <div>
+          <p className="font-medium text-text-main mb-0.5">Model Playground</p>
+          <p>
+            Test any model directly from the dashboard. Pick a provider, model, and endpoint type,
+            then send a request to see the raw response.
+          </p>
+        </div>
+      </div>
+
+      {/* Controls */}
+      <Card>
+        <div className="p-4 flex flex-col sm:flex-row items-end gap-4">
+          {/* Provider */}
+          <div className="flex-1 w-full">
+            <label className="block text-xs font-medium text-text-muted mb-1.5 uppercase tracking-wider">
+              Provider
+            </label>
+            <Select
+              value={selectedProvider}
+              onChange={(e: any) => handleProviderChange(e.target.value)}
+              options={providers}
+              className="w-full"
+            />
+          </div>
+
+          {/* Model */}
+          <div className="flex-1 w-full">
+            <label className="block text-xs font-medium text-text-muted mb-1.5 uppercase tracking-wider">
+              Model
+            </label>
+            <Select
+              value={selectedModel}
+              onChange={(e: any) => handleModelChange(e.target.value)}
+              options={filteredModels}
+              className="w-full"
+            />
+          </div>
+
+          {/* Endpoint */}
+          <div className="flex-1 w-full">
+            <label className="block text-xs font-medium text-text-muted mb-1.5 uppercase tracking-wider">
+              Endpoint
+            </label>
+            <Select
+              value={selectedEndpoint}
+              onChange={(e: any) => handleEndpointChange(e.target.value)}
+              options={ENDPOINT_OPTIONS}
+              className="w-full"
+            />
+          </div>
+
+          {/* Send Button */}
+          <div className="shrink-0">
+            {loading ? (
+              <Button icon="stop" variant="secondary" onClick={handleCancel}>
+                Cancel
+              </Button>
+            ) : (
+              <Button
+                icon="send"
+                onClick={handleSend}
+                disabled={!requestBody.trim() || !selectedModel}
+              >
+                Send
+              </Button>
+            )}
+          </div>
+        </div>
+      </Card>
+
+      {/* Split Editor View */}
+      <div className="grid grid-cols-1 lg:grid-cols-2 gap-4">
+        {/* Request Panel */}
+        <Card>
+          <div className="p-4 space-y-3">
+            <div className="flex items-center justify-between">
+              <div className="flex items-center gap-2">
+                <span className="material-symbols-outlined text-[18px] text-text-muted">
+                  upload
+                </span>
+                <h3 className="text-sm font-semibold text-text-main">Request</h3>
+                <Badge variant="info" size="sm">
+                  POST {ENDPOINT_PATHS[selectedEndpoint]}
+                </Badge>
+              </div>
+              <div className="flex items-center gap-1">
+                <button
+                  onClick={() => handleCopy(requestBody)}
+                  className="p-1.5 rounded hover:bg-black/5 dark:hover:bg-white/5 text-text-muted hover:text-text-main transition-colors"
+                  title="Copy"
+                >
+                  <span className="material-symbols-outlined text-[16px]">content_copy</span>
+                </button>
+                <button
+                  onClick={() => {
+                    const template = { ...DEFAULT_BODIES[selectedEndpoint] };
+                    if ("model" in template) (template as any).model = selectedModel;
+                    setRequestBody(JSON.stringify(template, null, 2));
+                  }}
+                  className="p-1.5 rounded hover:bg-black/5 dark:hover:bg-white/5 text-text-muted hover:text-text-main transition-colors"
+                  title="Reset to default"
+                >
+                  <span className="material-symbols-outlined text-[16px]">restart_alt</span>
+                </button>
+              </div>
+            </div>
+            <div className="border border-border rounded-lg overflow-hidden">
+              <Editor
+                height="400px"
+                defaultLanguage="json"
+                value={requestBody}
+                onChange={(value: string | undefined) => setRequestBody(value || "")}
+                theme="vs-dark"
+                options={{
+                  minimap: { enabled: false },
+                  fontSize: 12,
+                  lineNumbers: "on",
+                  scrollBeyondLastLine: false,
+                  wordWrap: "on",
+                  automaticLayout: true,
+                  formatOnPaste: true,
+                }}
+              />
+            </div>
+          </div>
+        </Card>
+
+        {/* Response Panel */}
+        <Card>
+          <div className="p-4 space-y-3">
+            <div className="flex items-center justify-between">
+              <div className="flex items-center gap-2">
+                <span className="material-symbols-outlined text-[18px] text-text-muted">
+                  download
+                </span>
+                <h3 className="text-sm font-semibold text-text-main">Response</h3>
+                {responseStatus !== null && (
+                  <Badge
+                    variant={responseStatus >= 200 && responseStatus < 300 ? "success" : "error"}
+                    size="sm"
+                  >
+                    {responseStatus}
+                  </Badge>
+                )}
+                {responseDuration !== null && (
+                  <span className="text-xs text-text-muted">{responseDuration}ms</span>
+                )}
+                {loading && (
+                  <span className="material-symbols-outlined text-[14px] text-primary animate-spin">
+                    progress_activity
+                  </span>
+                )}
+              </div>
+              <div className="flex items-center gap-1">
+                <button
+                  onClick={() => handleCopy(responseBody)}
+                  className="p-1.5 rounded hover:bg-black/5 dark:hover:bg-white/5 text-text-muted hover:text-text-main transition-colors"
+                  title="Copy"
+                >
+                  <span className="material-symbols-outlined text-[16px]">content_copy</span>
+                </button>
+              </div>
+            </div>
+            <div className="border border-border rounded-lg overflow-hidden">
+              <Editor
+                height="400px"
+                defaultLanguage="json"
+                value={responseBody}
+                theme="vs-dark"
+                options={{
+                  minimap: { enabled: false },
+                  fontSize: 12,
+                  lineNumbers: "on",
+                  scrollBeyondLastLine: false,
+                  wordWrap: "on",
+                  automaticLayout: true,
+                  readOnly: true,
+                }}
+              />
+            </div>
+          </div>
+        </Card>
+      </div>
+    </div>
+  );
+}
@@ -1338,6 +1338,8 @@ function CustomModelsSection({ providerId, providerAlias, copied, onCopy }) {
  const [customModels, setCustomModels] = useState([]);
  const [newModelId, setNewModelId] = useState("");
  const [newModelName, setNewModelName] = useState("");
+  const [newApiFormat, setNewApiFormat] = useState("chat-completions");
+  const [newEndpoints, setNewEndpoints] = useState(["chat"]);
  const [adding, setAdding] = useState(false);
  const [loading, setLoading] = useState(true);

@@ -1370,11 +1372,15 @@ function CustomModelsSection({ providerId, providerAlias, copied, onCopy }) {
          provider: providerId,
          modelId: newModelId.trim(),
          modelName: newModelName.trim() || undefined,
+          apiFormat: newApiFormat,
+          supportedEndpoints: newEndpoints,
        }),
      });
      if (res.ok) {
        setNewModelId("");
        setNewModelName("");
+        setNewApiFormat("chat-completions");
+        setNewEndpoints(["chat"]);
        await fetchCustomModels();
      }
    } catch (e) {
@@ -1407,38 +1413,89 @@ function CustomModelsSection({ providerId, providerAlias, copied, onCopy }) {
      <p className="text-xs text-text-muted mb-3">{t("customModelsHint")}</p>

      {/* Add form */}
-      <div className="flex items-end gap-2 mb-3">
-        <div className="flex-1">
-          <label htmlFor="custom-model-id" className="text-xs text-text-muted mb-1 block">
-            {t("modelId")}
-          </label>
-          <input
-            id="custom-model-id"
-            type="text"
-            value={newModelId}
-            onChange={(e) => setNewModelId(e.target.value)}
-            onKeyDown={(e) => e.key === "Enter" && handleAdd()}
-            placeholder={t("customModelPlaceholder")}
-            className="w-full px-3 py-2 text-sm border border-border rounded-lg bg-background focus:outline-none focus:border-primary"
-          />
+      <div className="flex flex-col gap-3 mb-3">
+        <div className="flex items-end gap-2">
+          <div className="flex-1">
+            <label htmlFor="custom-model-id" className="text-xs text-text-muted mb-1 block">
+              {t("modelId")}
+            </label>
+            <input
+              id="custom-model-id"
+              type="text"
+              value={newModelId}
+              onChange={(e) => setNewModelId(e.target.value)}
+              onKeyDown={(e) => e.key === "Enter" && handleAdd()}
+              placeholder={t("customModelPlaceholder")}
+              className="w-full px-3 py-2 text-sm border border-border rounded-lg bg-background focus:outline-none focus:border-primary"
+            />
+          </div>
+          <div className="w-40">
+            <label htmlFor="custom-model-name" className="text-xs text-text-muted mb-1 block">
+              {t("displayName")}
+            </label>
+            <input
+              id="custom-model-name"
+              type="text"
+              value={newModelName}
+              onChange={(e) => setNewModelName(e.target.value)}
+              onKeyDown={(e) => e.key === "Enter" && handleAdd()}
+              placeholder={t("optional")}
+              className="w-full px-3 py-2 text-sm border border-border rounded-lg bg-background focus:outline-none focus:border-primary"
+            />
+          </div>
+          <Button size="sm" icon="add" onClick={handleAdd} disabled={!newModelId.trim() || adding}>
+            {adding ? t("adding") : t("add")}
+          </Button>
        </div>
-        <div className="w-40">
-          <label htmlFor="custom-model-name" className="text-xs text-text-muted mb-1 block">
-            {t("displayName")}
-          </label>
-          <input
-            id="custom-model-name"
-            type="text"
-            value={newModelName}
-            onChange={(e) => setNewModelName(e.target.value)}
-            onKeyDown={(e) => e.key === "Enter" && handleAdd()}
-            placeholder={t("optional")}
-            className="w-full px-3 py-2 text-sm border border-border rounded-lg bg-background focus:outline-none focus:border-primary"
-          />
+
+        {/* API Format + Supported Endpoints */}
+        <div className="flex items-end gap-4 flex-wrap">
+          <div className="w-48">
+            <label htmlFor="custom-api-format" className="text-xs text-text-muted mb-1 block">
+              API Format
+            </label>
+            <select
+              id="custom-api-format"
+              value={newApiFormat}
+              onChange={(e) => setNewApiFormat(e.target.value)}
+              className="w-full px-3 py-2 text-sm border border-border rounded-lg bg-background focus:outline-none focus:border-primary"
+            >
+              <option value="chat-completions">Chat Completions</option>
+              <option value="responses">Responses API</option>
+            </select>
+          </div>
+          <div className="flex-1">
+            <span className="text-xs text-text-muted mb-1 block">Supported Endpoints</span>
+            <div className="flex items-center gap-3">
+              {["chat", "embeddings", "images", "audio"].map((ep) => (
+                <label
+                  key={ep}
+                  className="flex items-center gap-1.5 text-xs text-text-main cursor-pointer"
+                >
+                  <input
+                    type="checkbox"
+                    checked={newEndpoints.includes(ep)}
+                    onChange={(e) => {
+                      if (e.target.checked) {
+                        setNewEndpoints((prev) => [...prev, ep]);
+                      } else {
+                        setNewEndpoints((prev) => prev.filter((x) => x !== ep));
+                      }
+                    }}
+                    className="rounded border-border"
+                  />
+                  {ep === "chat"
+                    ? "💬 Chat"
+                    : ep === "embeddings"
+                      ? "📐 Embeddings"
+                      : ep === "images"
+                        ? "🖼️ Images"
+                        : "🔊 Audio"}
+                </label>
+              ))}
+            </div>
+          </div>
        </div>
-        <Button size="sm" icon="add" onClick={handleAdd} disabled={!newModelId.trim() || adding}>
-          {adding ? t("adding") : t("add")}
-        </Button>
      </div>

      {/* List */}
@@ -1457,7 +1514,7 @@ function CustomModelsSection({ providerId, providerAlias, copied, onCopy }) {
                <span className="material-symbols-outlined text-base text-primary">tune</span>
                <div className="flex-1 min-w-0">
                  <p className="text-sm font-medium truncate">{model.name || model.id}</p>
-                  <div className="flex items-center gap-1 mt-1">
+                  <div className="flex items-center gap-1 mt-1 flex-wrap">
                    <code className="text-xs text-text-muted font-mono bg-sidebar px-1.5 py-0.5 rounded">
                      {fullModel}
                    </code>
@@ -1470,6 +1527,26 @@ function CustomModelsSection({ providerId, providerAlias, copied, onCopy }) {
                        {copied === copyKey ? "check" : "content_copy"}
                      </span>
                    </button>
+                    {model.apiFormat === "responses" && (
+                      <span className="text-[10px] px-1.5 py-0.5 rounded-full bg-blue-500/15 text-blue-400 font-medium">
+                        Responses
+                      </span>
+                    )}
+                    {model.supportedEndpoints?.includes("embeddings") && (
+                      <span className="text-[10px] px-1.5 py-0.5 rounded-full bg-purple-500/15 text-purple-400 font-medium">
+                        📐 Embed
+                      </span>
+                    )}
+                    {model.supportedEndpoints?.includes("images") && (
+                      <span className="text-[10px] px-1.5 py-0.5 rounded-full bg-amber-500/15 text-amber-400 font-medium">
+                        🖼️ Images
+                      </span>
+                    )}
+                    {model.supportedEndpoints?.includes("audio") && (
+                      <span className="text-[10px] px-1.5 py-0.5 rounded-full bg-green-500/15 text-green-400 font-medium">
+                        🔊 Audio
+                      </span>
+                    )}
                  </div>
                </div>
                <button
@@ -248,6 +248,76 @@ export default function SecurityTab() {
        </div>
      </Card>

+      {/* CLI Fingerprint Matching */}
+      <Card>
+        <div className="flex items-center gap-3 mb-4">
+          <div className="p-2 rounded-lg bg-emerald-500/10 text-emerald-600 dark:text-emerald-400">
+            <span className="material-symbols-outlined text-[20px]" aria-hidden="true">
+              fingerprint
+            </span>
+          </div>
+          <h3 className="text-lg font-semibold">{t("cliFingerprint")}</h3>
+        </div>
+        <div className="flex flex-col gap-4">
+          <p className="text-sm text-text-muted">{t("cliFingerprintDesc")}</p>
+          <div className="flex flex-wrap gap-2">
+            {(["codex", "claude", "github", "antigravity"] as const).map((providerId) => {
+              const providerMeta = Object.values(AI_PROVIDERS).find(
+                (p: any) => p.id === providerId
+              ) as any;
+              const isEnabled = (settings.cliCompatProviders || []).includes(providerId);
+              const displayName = providerMeta?.name || providerId;
+              const icon = providerMeta?.icon || "terminal";
+              const color = providerMeta?.color || "#888";
+              return (
+                <button
+                  key={providerId}
+                  onClick={() => {
+                    const current: string[] = settings.cliCompatProviders || [];
+                    const updated = current.includes(providerId)
+                      ? current.filter((p) => p !== providerId)
+                      : [...current, providerId];
+                    updateSetting("cliCompatProviders", updated);
+                  }}
+                  disabled={loading}
+                  className={`inline-flex items-center gap-1.5 px-2.5 py-1.5 rounded-lg text-xs font-medium transition-all border ${
+                    isEnabled
+                      ? "bg-emerald-500/10 border-emerald-500/30 text-emerald-600 dark:text-emerald-400"
+                      : "bg-black/[0.02] dark:bg-white/[0.02] border-transparent text-text-muted hover:bg-black/[0.05] dark:hover:bg-white/[0.05]"
+                  }`}
+                  title={
+                    isEnabled
+                      ? t("disableFingerprintTitle", { provider: displayName })
+                      : t("enableFingerprintTitle", { provider: displayName })
+                  }
+                >
+                  <span
+                    className="material-symbols-outlined text-[14px]"
+                    style={{ color: isEnabled ? undefined : color }}
+                  >
+                    {isEnabled ? "fingerprint" : icon}
+                  </span>
+                  {displayName}
+                  {isEnabled && (
+                    <span className="material-symbols-outlined text-[12px] text-emerald-500">
+                      check
+                    </span>
+                  )}
+                </button>
+              );
+            })}
+          </div>
+          {(settings.cliCompatProviders || []).length > 0 && (
+            <p className="text-xs text-emerald-600 dark:text-emerald-400 mt-1 flex items-center gap-1">
+              <span className="material-symbols-outlined text-[14px]">verified</span>
+              {t("cliFingerprintEnabled", {
+                count: (settings.cliCompatProviders || []).length,
+              })}
+            </p>
+          )}
+        </div>
+      </Card>
+
      <SessionInfoCard />
      <IPFilterSection />
    </div>
@@ -150,10 +150,9 @@ export default function ProviderLimitCard({
      {!loading && !error && !message && quotas?.length > 0 && (
        <div className="space-y-4">
          {quotas.map((quota, index) => {
-            // For Antigravity, use remainingPercentage if available, otherwise calculate
            const percentage =
              quota.remainingPercentage !== undefined
-                ? Math.round(((quota.total - quota.used) / quota.total) * 100)
+                ? Math.round(quota.remainingPercentage)
                : calculatePercentage(quota.used, quota.total);
            const unlimited = quota.total === 0 || quota.total === null;

@@ -1,4 +1,5 @@
 import { getModelsByProviderId } from "@omniroute/open-sse/config/providerModels.ts";
+import { safePercentage } from "@/shared/utils/formatting";

 /**
 * Format ISO date string to countdown format (inspired by vscode-antigravity-cockpit)
@@ -110,7 +111,7 @@ export function parseQuotaData(provider, data) {
              used: quota.used || 0,
              total: quota.total || 0,
              resetAt: quota.resetAt || null,
-              remainingPercentage: quota.remainingPercentage,
+              remainingPercentage: safePercentage(quota.remainingPercentage),
            });
          });
        }
@@ -159,7 +160,7 @@ export function parseQuotaData(provider, data) {
              used: quota.used || 0,
              total: quota.total || 0,
              resetAt: quota.resetAt || null,
-              remainingPercentage: quota.remainingPercentage,
+              remainingPercentage: safePercentage(quota.remainingPercentage),
            });
          });
        }
@@ -0,0 +1,35 @@
+/**
+ * API Route: /api/acp/agents
+ *
+ * Returns the list of detected CLI agents and their availability status.
+ * Used by the dashboard to show ACP transport options.
+ */
+
+import { NextResponse } from "next/server";
+import { detectInstalledAgents } from "@/lib/acp";
+
+export const dynamic = "force-dynamic";
+
+export async function GET() {
+  try {
+    const agents = detectInstalledAgents();
+    return NextResponse.json({
+      agents: agents.map((a) => ({
+        id: a.id,
+        name: a.name,
+        binary: a.binary,
+        version: a.version,
+        installed: a.installed,
+        providerAlias: a.providerAlias,
+        protocol: a.protocol,
+      })),
+      available: agents.filter((a) => a.installed).length,
+      total: agents.length,
+    });
+  } catch (error: any) {
+    return NextResponse.json(
+      { error: error.message || "Failed to detect agents" },
+      { status: 500 }
+    );
+  }
+}
@@ -221,9 +221,15 @@ export async function POST(
      let connection: any;
      if (tokenData.email) {
        const existing = await getProviderConnections({ provider });
-        const match = existing.find(
-          (c: any) => c.email === tokenData.email && c.authType === "oauth"
-        );
+        const match = existing.find((c: any) => {
+          if (c.email !== tokenData.email || c.authType !== "oauth") return false;
+          // For Codex, also check workspaceId to avoid overwriting different workspace connections
+          if (provider === "codex" && tokenData.providerSpecificData?.workspaceId) {
+            const existingWorkspace = c.providerSpecificData?.workspaceId;
+            return existingWorkspace === tokenData.providerSpecificData.workspaceId;
+          }
+          return true;
+        });
        const matchId = typeof match?.id === "string" ? match.id : null;
        if (matchId) {
          connection = await updateProviderConnection(matchId, {
@@ -285,9 +291,15 @@ export async function POST(
        let connection: any;
        if (result.tokens.email) {
          const existing = await getProviderConnections({ provider });
-          const match = existing.find(
-            (c: any) => c.email === result.tokens.email && c.authType === "oauth"
-          );
+          const match = existing.find((c: any) => {
+            if (c.email !== result.tokens.email || c.authType !== "oauth") return false;
+            // For Codex, also check workspaceId to avoid overwriting different workspace connections
+            if (provider === "codex" && result.tokens.providerSpecificData?.workspaceId) {
+              const existingWorkspace = c.providerSpecificData?.workspaceId;
+              return existingWorkspace === result.tokens.providerSpecificData.workspaceId;
+            }
+            return true;
+          });
          const matchId = typeof match?.id === "string" ? match.id : null;
          if (matchId) {
            connection = await updateProviderConnection(matchId, {
@@ -399,9 +411,15 @@ export async function POST(
        let connection: any;
        if (tokenData.email) {
          const existing = await getProviderConnections({ provider });
-          const match = existing.find(
-            (c: any) => c.email === tokenData.email && c.authType === "oauth"
-          );
+          const match = existing.find((c: any) => {
+            if (c.email !== tokenData.email || c.authType !== "oauth") return false;
+            // For Codex, also check workspaceId to avoid overwriting different workspace connections
+            if (provider === "codex" && tokenData.providerSpecificData?.workspaceId) {
+              const existingWorkspace = c.providerSpecificData?.workspaceId;
+              return existingWorkspace === tokenData.providerSpecificData.workspaceId;
+            }
+            return true;
+          });
          const matchId = typeof match?.id === "string" ? match.id : null;
          if (matchId) {
            connection = await updateProviderConnection(matchId, {
@@ -64,9 +64,16 @@ export async function POST(request) {
    if (isValidationFailure(validation)) {
      return Response.json({ error: validation.error }, { status: 400 });
    }
-    const { provider, modelId, modelName, source } = validation.data;
+    const { provider, modelId, modelName, source, apiFormat, supportedEndpoints } = validation.data;

-    const model = await addCustomModel(provider, modelId, modelName, source || "manual");
+    const model = await addCustomModel(
+      provider,
+      modelId,
+      modelName,
+      source || "manual",
+      apiFormat,
+      supportedEndpoints
+    );
    return Response.json({ model });
  } catch (error) {
    console.error("Error adding provider model:", error);
@@ -5,12 +5,18 @@ import bcrypt from "bcryptjs";
 import { getRuntimePorts } from "@/lib/runtime/ports";
 import { updateSettingsSchema } from "@/shared/validation/settingsSchemas";
 import { isValidationFailure, validateBody } from "@/shared/validation/helpers";
+import { setCliCompatProviders } from "../../../../open-sse/config/cliFingerprints";

 export async function GET() {
  try {
    const settings = await getSettings();
    const { password, ...safeSettings } = settings;

+    // Sync CLI fingerprint providers to runtime cache on load
+    if (settings.cliCompatProviders) {
+      setCliCompatProviders(settings.cliCompatProviders as string[]);
+    }
+
    const enableRequestLogs = process.env.ENABLE_REQUEST_LOGS === "true";
    const runtimePorts = getRuntimePorts();

@@ -74,6 +80,11 @@ export async function PATCH(request) {
      clearHealthCheckLogCache();
    }

+    // Sync CLI fingerprint providers to runtime cache
+    if ("cliCompatProviders" in body) {
+      setCliCompatProviders(body.cliCompatProviders || []);
+    }
+
    const { password, ...safeSettings } = settings;
    return NextResponse.json(safeSettings);
  } catch (error) {
@@ -4,6 +4,11 @@ import { getUsageForProvider } from "@omniroute/open-sse/services/usage.ts";
 import { getExecutor } from "@omniroute/open-sse/executors/index.ts";
 import { syncToCloud } from "@/lib/cloudSync";
 import { runWithProxyContext } from "@omniroute/open-sse/utils/proxyFetch.ts";
+import { setQuotaCache } from "@/domain/quotaCache";
+
+function isRecord(value: unknown): value is Record<string, any> {
+  return value !== null && typeof value === "object" && !Array.isArray(value);
+}

 /**
 * Sync to cloud if enabled
@@ -147,6 +152,12 @@ export async function GET(request: Request, { params }: { params: Promise<{ conn
    const usage = await runWithProxyContext(proxyInfo?.proxy || null, () =>
      getUsageForProvider(connection)
    );
+
+    // Populate quota cache for quota-aware account selection
+    if (isRecord(usage?.quotas)) {
+      setQuotaCache(connectionId, connection.provider, usage.quotas);
+    }
+
    return Response.json(usage);
  } catch (error) {
    console.error("[Usage API] Error fetching usage:", error);
@@ -13,6 +13,8 @@ import { enforceApiKeyPolicy } from "@/shared/utils/apiKeyPolicy";
 import { v1EmbeddingsSchema } from "@/shared/validation/schemas";
 import { isValidationFailure, validateBody } from "@/shared/validation/helpers";

+import { getAllCustomModels } from "@/lib/localDb";
+
 /**
 * Handle CORS preflight
 */
@@ -30,23 +32,43 @@ export async function OPTIONS() {
 * GET /v1/embeddings — list available embedding models
 */
 export async function GET() {
-  const models = getAllEmbeddingModels();
-  return new Response(
-    JSON.stringify({
-      object: "list",
-      data: models.map((m) => ({
-        id: m.id,
-        object: "model",
-        created: Math.floor(Date.now() / 1000),
-        owned_by: m.provider,
-        type: "embedding",
-        dimensions: m.dimensions,
-      })),
-    }),
-    {
-      headers: { "Content-Type": "application/json" },
+  const builtInModels = getAllEmbeddingModels();
+  const timestamp = Math.floor(Date.now() / 1000);
+
+  const data = builtInModels.map((m) => ({
+    id: m.id,
+    object: "model",
+    created: timestamp,
+    owned_by: m.provider,
+    type: "embedding",
+    dimensions: m.dimensions,
+  }));
+
+  // Include custom models tagged for embeddings
+  try {
+    const customModelsMap = (await getAllCustomModels()) as Record<string, any>;
+    for (const [providerId, models] of Object.entries(customModelsMap)) {
+      if (!Array.isArray(models)) continue;
+      for (const model of models) {
+        if (!model?.id || !Array.isArray(model.supportedEndpoints)) continue;
+        if (!model.supportedEndpoints.includes("embeddings")) continue;
+        const fullId = `${providerId}/${model.id}`;
+        if (data.some((d) => d.id === fullId)) continue;
+        data.push({
+          id: fullId,
+          object: "model",
+          created: timestamp,
+          owned_by: providerId,
+          type: "embedding",
+          dimensions: null,
+        });
+      }
    }
-  );
+  } catch {}
+
+  return new Response(JSON.stringify({ object: "list", data }), {
+    headers: { "Content-Type": "application/json" },
+  });
 }

 /**
@@ -14,6 +14,8 @@ import { enforceApiKeyPolicy } from "@/shared/utils/apiKeyPolicy";
 import { v1ImageGenerationSchema } from "@/shared/validation/schemas";
 import { isValidationFailure, validateBody } from "@/shared/validation/helpers";

+import { getAllCustomModels } from "@/lib/localDb";
+
 /**
 * Handle CORS preflight
 */
@@ -31,23 +33,43 @@ export async function OPTIONS() {
 * GET /v1/images/generations — list available image models
 */
 export async function GET() {
-  const models = getAllImageModels();
-  return new Response(
-    JSON.stringify({
-      object: "list",
-      data: models.map((m) => ({
-        id: m.id,
-        object: "model",
-        created: Math.floor(Date.now() / 1000),
-        owned_by: m.provider,
-        type: "image",
-        supported_sizes: m.supportedSizes,
-      })),
-    }),
-    {
-      headers: { "Content-Type": "application/json" },
+  const builtInModels = getAllImageModels();
+  const timestamp = Math.floor(Date.now() / 1000);
+
+  const data = builtInModels.map((m) => ({
+    id: m.id,
+    object: "model",
+    created: timestamp,
+    owned_by: m.provider,
+    type: "image",
+    supported_sizes: m.supportedSizes,
+  }));
+
+  // Include custom models tagged for images
+  try {
+    const customModelsMap = (await getAllCustomModels()) as Record<string, any>;
+    for (const [providerId, models] of Object.entries(customModelsMap)) {
+      if (!Array.isArray(models)) continue;
+      for (const model of models) {
+        if (!model?.id || !Array.isArray(model.supportedEndpoints)) continue;
+        if (!model.supportedEndpoints.includes("images")) continue;
+        const fullId = `${providerId}/${model.id}`;
+        if (data.some((d) => d.id === fullId)) continue;
+        data.push({
+          id: fullId,
+          object: "model",
+          created: timestamp,
+          owned_by: providerId,
+          type: "image",
+          supported_sizes: null,
+        });
+      }
    }
-  );
+  } catch {}
+
+  return new Response(JSON.stringify({ object: "list", data }), {
+    headers: { "Content-Type": "application/json" },
+  });
 }

 /**
@@ -85,7 +107,30 @@ export async function POST(request) {
  if (policy.rejection) return policy.rejection;

  // Parse model to get provider
-  const { provider } = parseImageModel(body.model);
+  let { provider } = parseImageModel(body.model);
+  let isCustomModel = false;
+
+  // If not in built-in registry, check custom models tagged for images
+  if (!provider) {
+    try {
+      const customModelsMap = (await getAllCustomModels()) as Record<string, any>;
+      for (const [providerId, models] of Object.entries(customModelsMap)) {
+        if (!Array.isArray(models)) continue;
+        for (const model of models) {
+          if (!model?.id || !Array.isArray(model.supportedEndpoints)) continue;
+          if (!model.supportedEndpoints.includes("images")) continue;
+          const fullId = `${providerId}/${model.id}`;
+          if (fullId === body.model) {
+            provider = providerId;
+            isCustomModel = true;
+            break;
+          }
+        }
+        if (provider) break;
+      }
+    } catch {}
+  }
+
  if (!provider) {
    return errorResponse(
      HTTP_STATUS.BAD_REQUEST,
@@ -106,9 +151,23 @@ export async function POST(request) {
        `No credentials for image provider: ${provider}`
      );
    }
+  } else if (isCustomModel) {
+    // Custom models need credentials from the provider connection
+    credentials = await getProviderCredentials(provider);
+    if (!credentials) {
+      return errorResponse(
+        HTTP_STATUS.BAD_REQUEST,
+        `No credentials for custom image provider: ${provider}`
+      );
+    }
  }

-  const result = await handleImageGeneration({ body, credentials, log });
+  const result = await handleImageGeneration({
+    body,
+    credentials,
+    log,
+    ...(isCustomModel && { resolvedProvider: provider }),
+  });

  if (result.success) {
    return new Response(JSON.stringify((result as any).data), {
@@ -364,6 +364,17 @@ export async function getUnifiedModelsResponse(
          const aliasId = `${alias}/${modelId}`;
          if (models.some((m) => m.id === aliasId)) continue;

+          // Determine type from supportedEndpoints
+          const endpoints = Array.isArray(model.supportedEndpoints)
+            ? model.supportedEndpoints
+            : ["chat"];
+          const apiFormat =
+            typeof model.apiFormat === "string" ? model.apiFormat : "chat-completions";
+          let modelType: string | undefined;
+          if (endpoints.includes("embeddings")) modelType = "embedding";
+          else if (endpoints.includes("images")) modelType = "image";
+          else if (endpoints.includes("audio")) modelType = "audio";
+
          models.push({
            id: aliasId,
            object: "model",
@@ -373,6 +384,11 @@ export async function getUnifiedModelsResponse(
            root: modelId,
            parent: null,
            custom: true,
+            ...(modelType ? { type: modelType } : {}),
+            ...(apiFormat !== "chat-completions" ? { api_format: apiFormat } : {}),
+            ...(endpoints.length > 1 || !endpoints.includes("chat")
+              ? { supported_endpoints: endpoints }
+              : {}),
          });

          // Only add provider-prefixed version if different from alias
@@ -388,6 +404,7 @@ export async function getUnifiedModelsResponse(
              root: modelId,
              parent: aliasId,
              custom: true,
+              ...(modelType ? { type: modelType } : {}),
            });
          }
        }
@@ -6,6 +6,9 @@
   directives ensure all utility classes in route groups are included. */
@source "../app/(dashboard)";
@source "../../open-sse";
+@source not "../../*.sqlite*";
+@source not "../../.claude*";
+@source not "../../.claude-memory";

@custom-variant dark (&:where(.dark, .dark *));

@@ -0,0 +1,264 @@
+/**
+ * Quota Cache — Domain Layer
+ *
+ * In-memory cache of provider quota data per connectionId.
+ * Populated by:
+ *   - Dashboard usage endpoint (GET /api/usage/[connectionId])
+ *   - 429 responses marking account as exhausted
+ *
+ * Background refresh runs every 1 minute:
+ *   - Active accounts (quota > 0%): refetch every 5 minutes
+ *   - Exhausted accounts: refetch every 5 minutes (or immediately after resetAt passes)
+ *
+ * @module domain/quotaCache
+ */
+
+import { getUsageForProvider } from "@omniroute/open-sse/services/usage.ts";
+import { getProviderConnectionById, resolveProxyForConnection } from "@/lib/localDb";
+import { runWithProxyContext } from "@omniroute/open-sse/utils/proxyFetch.ts";
+import { safePercentage } from "@/shared/utils/formatting";
+
+// ─── Types ──────────────────────────────────────────────────────────────────
+
+interface QuotaInfo {
+  remainingPercentage: number;
+  resetAt: string | null;
+}
+
+interface QuotaCacheEntry {
+  connectionId: string;
+  provider: string;
+  quotas: Record<string, QuotaInfo>;
+  fetchedAt: number;
+  exhausted: boolean;
+  nextResetAt: string | null;
+}
+
+// ─── Constants ──────────────────────────────────────────────────────────────
+
+const ACTIVE_TTL_MS = 5 * 60 * 1000; // 5 minutes for active accounts
+const EXHAUSTED_TTL_MS = 5 * 60 * 1000; // 5 minutes for 429-sourced entries (no resetAt)
+const EXHAUSTED_REFRESH_MS = 5 * 60 * 1000; // 5 minutes: recheck exhausted accounts (aligned with TTL)
+const REFRESH_INTERVAL_MS = 60 * 1000; // Background tick every 1 minute
+
+// ─── State ──────────────────────────────────────────────────────────────────
+
+const cache = new Map<string, QuotaCacheEntry>();
+const MAX_CONCURRENT_REFRESHES = 5;
+let refreshTimer: ReturnType<typeof setInterval> | null = null;
+let tickRunning = false;
+
+// ─── Helpers ────────────────────────────────────────────────────────────────
+
+function isExhausted(quotas: Record<string, QuotaInfo>): boolean {
+  const entries = Object.values(quotas);
+  if (entries.length === 0) return false;
+  return entries.every((q) => q.remainingPercentage <= 0);
+}
+
+function parseDate(value: string): number | null {
+  const ms = new Date(value).getTime();
+  return Number.isNaN(ms) ? null : ms;
+}
+
+function earliestResetAt(quotas: Record<string, QuotaInfo>): string | null {
+  let earliest: string | null = null;
+  let earliestMs = Infinity;
+  for (const q of Object.values(quotas)) {
+    if (!q.resetAt) continue;
+    const ms = parseDate(q.resetAt);
+    if (ms !== null && ms < earliestMs) {
+      earliestMs = ms;
+      earliest = q.resetAt;
+    }
+  }
+  return earliest;
+}
+
+function normalizeQuotas(rawQuotas: Record<string, any>): Record<string, QuotaInfo> {
+  const result: Record<string, QuotaInfo> = {};
+  for (const [key, q] of Object.entries(rawQuotas)) {
+    if (q && typeof q === "object") {
+      result[key] = {
+        remainingPercentage:
+          safePercentage(q.remainingPercentage) ??
+          (q.total > 0 ? Math.round(((q.total - (q.used || 0)) / q.total) * 100) : 0),
+        resetAt: q.resetAt || null,
+      };
+    }
+  }
+  return result;
+}
+
+// ─── Public API ─────────────────────────────────────────────────────────────
+
+/**
+ * Store quota data for a connection (called by usage endpoint and background refresh).
+ */
+export function setQuotaCache(
+  connectionId: string,
+  provider: string,
+  rawQuotas: Record<string, any>
+) {
+  const quotas = normalizeQuotas(rawQuotas);
+  const exhausted = isExhausted(quotas);
+  cache.set(connectionId, {
+    connectionId,
+    provider,
+    quotas,
+    fetchedAt: Date.now(),
+    exhausted,
+    nextResetAt: exhausted ? earliestResetAt(quotas) : null,
+  });
+}
+
+/**
+ * Get cached quota entry (returns null if not cached).
+ */
+export function getQuotaCache(connectionId: string): QuotaCacheEntry | null {
+  return cache.get(connectionId) || null;
+}
+
+/**
+ * Check if an account's quota is exhausted based on cached data.
+ * Returns false if no cache entry exists (unknown = assume available).
+ */
+export function isAccountQuotaExhausted(connectionId: string): boolean {
+  const entry = cache.get(connectionId);
+  if (!entry) return false;
+  if (!entry.exhausted) return false;
+
+  // If resetAt has passed, assume available until refresh confirms
+  if (entry.nextResetAt) {
+    const resetMs = parseDate(entry.nextResetAt);
+    if (resetMs !== null && resetMs <= Date.now()) return false;
+  }
+
+  // Exhausted entries without resetAt expire after fixed TTL
+  const age = Date.now() - entry.fetchedAt;
+  if (!entry.nextResetAt && age > EXHAUSTED_TTL_MS) return false;
+
+  return true;
+}
+
+/**
+ * Mark an account as quota-exhausted from a 429 response (no quota data available).
+ * Uses 5-minute fixed TTL since we don't know the actual resetAt.
+ */
+export function markAccountExhaustedFrom429(connectionId: string, provider: string) {
+  cache.set(connectionId, {
+    connectionId,
+    provider,
+    quotas: {},
+    fetchedAt: Date.now(),
+    exhausted: true,
+    nextResetAt: null,
+  });
+}
+
+// ─── Background Refresh ─────────────────────────────────────────────────────
+
+const refreshingSet = new Set<string>();
+
+async function refreshEntry(entry: QuotaCacheEntry) {
+  if (refreshingSet.has(entry.connectionId)) return;
+  refreshingSet.add(entry.connectionId);
+
+  try {
+    const connection = await getProviderConnectionById(entry.connectionId);
+    if (!connection || connection.authType !== "oauth" || !connection.isActive) {
+      cache.delete(entry.connectionId);
+      return;
+    }
+
+    const proxyInfo = await resolveProxyForConnection(entry.connectionId);
+    const usage = await runWithProxyContext(proxyInfo?.proxy || null, () =>
+      getUsageForProvider(connection)
+    );
+
+    if (usage?.quotas) {
+      setQuotaCache(entry.connectionId, entry.provider, usage.quotas);
+    }
+  } catch (err) {
+    console.warn(
+      `[QuotaCache] Refresh failed for ${entry.connectionId.slice(0, 8)}:`,
+      (err as any)?.message || err
+    );
+  } finally {
+    refreshingSet.delete(entry.connectionId);
+  }
+}
+
+function needsRefresh(entry: QuotaCacheEntry, now: number): boolean {
+  const age = now - entry.fetchedAt;
+  if (entry.exhausted) {
+    if (entry.nextResetAt) {
+      const resetMs = parseDate(entry.nextResetAt);
+      if (resetMs !== null && resetMs <= now) return true;
+    }
+    return age >= EXHAUSTED_REFRESH_MS;
+  }
+  return age >= ACTIVE_TTL_MS;
+}
+
+async function backgroundRefreshTick() {
+  if (tickRunning) return;
+  tickRunning = true;
+
+  try {
+    const now = Date.now();
+    const pending = [...cache.values()].filter((e) => needsRefresh(e, now));
+
+    // Refresh in batches to avoid thundering herd
+    for (let i = 0; i < pending.length; i += MAX_CONCURRENT_REFRESHES) {
+      const batch = pending.slice(i, i + MAX_CONCURRENT_REFRESHES);
+      await Promise.allSettled(batch.map(refreshEntry));
+    }
+  } finally {
+    tickRunning = false;
+  }
+}
+
+/**
+ * Start the background refresh timer.
+ */
+export function startBackgroundRefresh() {
+  if (refreshTimer) return;
+  refreshTimer = setInterval(backgroundRefreshTick, REFRESH_INTERVAL_MS);
+  refreshTimer?.unref?.();
+}
+
+/**
+ * Stop the background refresh timer.
+ */
+export function stopBackgroundRefresh() {
+  if (refreshTimer) {
+    clearInterval(refreshTimer);
+    refreshTimer = null;
+  }
+}
+
+/**
+ * Get cache stats (for debugging/dashboard).
+ */
+export function getQuotaCacheStats() {
+  const entries: Array<{
+    connectionId: string;
+    provider: string;
+    exhausted: boolean;
+    nextResetAt: string | null;
+    ageMs: number;
+  }> = [];
+
+  for (const entry of cache.values()) {
+    entries.push({
+      connectionId: entry.connectionId.slice(0, 8) + "...",
+      provider: entry.provider,
+      exhausted: entry.exhausted,
+      nextResetAt: entry.nextResetAt,
+      ageMs: Date.now() - entry.fetchedAt,
+    });
+  }
+
+  return { total: cache.size, entries };
+}
@@ -100,7 +100,8 @@
    "themeViolet": "بنفسجي ساحر",
    "themeOrange": "أورانج (Orange)",
    "themeCyan": "كيان",
-    "endpoints": "نقاط النهاية"
+    "endpoints": "نقاط النهاية",
+    "playground": "ملعب النماذج"
  },
  "themesPage": {
    "title": "المواضيع",
@@ -100,7 +100,8 @@
    "themeViolet": "Виолетово",
    "themeOrange": "Оранжево",
    "themeCyan": "Циан",
-    "endpoints": "Крайни точки"
+    "endpoints": "Крайни точки",
+    "playground": "Площадка"
  },
  "themesPage": {
    "title": "Теми",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Turkis",
-    "endpoints": "Endpoints"
+    "endpoints": "Endpoints",
+    "playground": "Legeplads"
  },
  "themesPage": {
    "title": "Temaer",
@@ -100,7 +100,8 @@
    "themeViolet": "Violett",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Endpunkte"
+    "endpoints": "Endpunkte",
+    "playground": "Spielwiese"
  },
  "themesPage": {
    "title": "Themen",
@@ -72,6 +72,7 @@
    "media": "Media",
    "settings": "Settings",
    "translator": "Translator",
+    "playground": "Playground",
    "docs": "Docs",
    "issues": "Issues",
    "endpoints": "Endpoints",
@@ -1492,6 +1493,11 @@
    "providersBlocked": "{count} provider(s) blocked from /models",
    "blockProviderTitle": "Block {provider}",
    "unblockProviderTitle": "Unblock {provider}",
+    "cliFingerprint": "CLI Fingerprint Matching",
+    "cliFingerprintDesc": "Match native CLI binary signatures when proxying requests. Reorders headers and body fields to look identical to the official CLI tools. Your proxy IP is preserved.",
+    "cliFingerprintEnabled": "{count} provider(s) with CLI fingerprint active",
+    "enableFingerprintTitle": "Enable fingerprint for {provider}",
+    "disableFingerprintTitle": "Disable fingerprint for {provider}",
    "routingStrategy": "Routing Strategy",
    "fillFirst": "Fill First",
    "fillFirstDesc": "Use accounts in priority order",
@@ -100,7 +100,8 @@
    "themeViolet": "Violeta",
    "themeOrange": "Naranja",
    "themeCyan": "cian",
-    "endpoints": "Endpoints"
+    "endpoints": "Endpoints",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Temas",
@@ -100,7 +100,8 @@
    "themeViolet": "Violetti",
    "themeOrange": "Oranssi",
    "themeCyan": "Syaani",
-    "endpoints": "Päätepisteet"
+    "endpoints": "Päätepisteet",
+    "playground": "Leikkipaikka"
  },
  "themesPage": {
    "title": "Teemat",
@@ -100,7 +100,8 @@
    "themeViolet": "Violette",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Points d'accès"
+    "endpoints": "Points d'accès",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Thèmes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "נקודות קצה"
+    "endpoints": "נקודות קצה",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "narancssárga",
    "themeCyan": "Cián",
-    "endpoints": "Végpontok"
+    "endpoints": "Végpontok",
+    "playground": "Játszótér"
  },
  "themesPage": {
    "title": "Témák",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Endpoint"
+    "endpoints": "Endpoint",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Endpoint"
+    "endpoints": "Endpoint",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Endpoint"
+    "endpoints": "Endpoint",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "エンドポイント"
+    "endpoints": "エンドポイント",
+    "playground": "プレイグラウンド"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "엔드포인트"
+    "endpoints": "엔드포인트",
+    "playground": "플레이그라운드"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Titik Akhir"
+    "endpoints": "Titik Akhir",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Eindpunten"
+    "endpoints": "Eindpunten",
+    "playground": "Speeltuin"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Endepunkter"
+    "endpoints": "Endepunkter",
+    "playground": "Lekeplass"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Mga Endpoint"
+    "endpoints": "Mga Endpoint",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Punkty końcowe"
+    "endpoints": "Punkty końcowe",
+    "playground": "Plac zabaw"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Endpoints"
+    "endpoints": "Endpoints",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -1480,6 +1481,11 @@
    "providersBlocked": "{count} provedor(es) bloqueado(s) do /models",
    "blockProviderTitle": "Bloquear {provider}",
    "unblockProviderTitle": "Desbloquear {provider}",
+    "cliFingerprint": "Assinatura Digital CLI",
+    "cliFingerprintDesc": "Reproduz a assinatura digital dos CLIs nativos ao fazer proxy. Reorganiza headers e campos do body para parecer idêntico às ferramentas CLI oficiais. O IP do proxy é preservado.",
+    "cliFingerprintEnabled": "{count} provider(s) com fingerprint CLI ativo",
+    "enableFingerprintTitle": "Ativar fingerprint para {provider}",
+    "disableFingerprintTitle": "Desativar fingerprint para {provider}",
    "routingStrategy": "Estratégia de Roteamento",
    "fillFirst": "Preencher Primeiro",
    "fillFirstDesc": "Usar contas em ordem de prioridade",
@@ -100,7 +100,8 @@
    "themeGreen": "Verde",
    "themeViolet": "Violeta",
    "themeOrange": "Laranja",
-    "themeCyan": "Ciano"
+    "themeCyan": "Ciano",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Puncte finale"
+    "endpoints": "Puncte finale",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Фиолетовый",
    "themeOrange": "Оранжевый",
    "themeCyan": "Голубой",
-    "endpoints": "Конечные точки"
+    "endpoints": "Конечные точки",
+    "playground": "Площадка"
  },
  "themesPage": {
    "title": "Темы",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Koncové body"
+    "endpoints": "Koncové body",
+    "playground": "Ihrisko"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Ändpunkter"
+    "endpoints": "Ändpunkter",
+    "playground": "Lekplats"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "จุดปลายทาง"
+    "endpoints": "จุดปลายทาง",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "ธีมส์",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Кінцеві точки"
+    "endpoints": "Кінцеві точки",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "Điểm cuối"
+    "endpoints": "Điểm cuối",
+    "playground": "Playground"
  },
  "themesPage": {
    "title": "Themes",
@@ -100,7 +100,8 @@
    "themeViolet": "Violet",
    "themeOrange": "Orange",
    "themeCyan": "Cyan",
-    "endpoints": "端点"
+    "endpoints": "端点",
+    "playground": "模型试验场"
  },
  "themesPage": {
    "title": "Themes",
@@ -8,22 +8,27 @@
 * @see https://nextjs.org/docs/app/building-your-application/optimizing/instrumentation
 */

-function ensureJwtSecret(): void {
+function ensureSecrets(): void {
+  // eslint-disable-next-line no-eval
+  const crypto = eval("require")("crypto");
+
  if (!process.env.JWT_SECRET || process.env.JWT_SECRET.trim() === "") {
-    // Use eval to hide require from webpack's static analysis
-    // This code only runs in Node.js runtime (guarded by NEXT_RUNTIME check)
-    // eslint-disable-next-line no-eval
-    const crypto = eval("require")("crypto");
    const generated = crypto.randomBytes(48).toString("base64");
    process.env.JWT_SECRET = generated;
    console.log("[STARTUP] JWT_SECRET auto-generated (random 64-char secret)");
  }
+
+  if (!process.env.API_KEY_SECRET || process.env.API_KEY_SECRET.trim() === "") {
+    const generated = crypto.randomBytes(32).toString("hex");
+    process.env.API_KEY_SECRET = generated;
+    console.log("[STARTUP] API_KEY_SECRET auto-generated (random 64-char hex secret)");
+  }
 }

 export async function register() {
  // Only run on the server (not during build or in Edge runtime)
  if (process.env.NEXT_RUNTIME === "nodejs") {
-    ensureJwtSecret();
+    ensureSecrets();
    // Console log file capture (must be first — before any logging occurs)
    const { initConsoleInterceptor } = await import("@/lib/consoleInterceptor");
    initConsoleInterceptor();
@@ -34,6 +39,13 @@ export async function register() {
    const { initApiBridgeServer } = await import("@/lib/apiBridgeServer");
    initApiBridgeServer();

+    // Quota cache: start background refresh for quota-aware account selection
+    // Dynamic import required — quotaCache depends on better-sqlite3 (Node-only),
+    // and instrumentation.ts is bundled for all runtimes including Edge.
+    const { startBackgroundRefresh } = await import("@/domain/quotaCache");
+    startBackgroundRefresh();
+    console.log("[STARTUP] Quota cache background refresh started");
+
    // Compliance: Initialize audit_log table + cleanup expired logs
    try {
      const { initAuditLog, cleanupExpiredLogs } = await import("@/lib/compliance/index");
@@ -0,0 +1,11 @@
+/**
+ * ACP Module — Public API
+ *
+ * Re-exports the registry and manager for convenient imports.
+ */
+
+export { detectInstalledAgents, getAgentById, getAvailableAgents } from "./registry";
+export type { CliAgentInfo } from "./registry";
+
+export { AcpManager, acpManager } from "./manager";
+export type { AcpSession } from "./manager";
@@ -0,0 +1,197 @@
+/**
+ * ACP (Agent Client Protocol) — Process Spawner & Manager
+ *
+ * Spawns CLI agents as child processes and manages their lifecycle.
+ * Communication happens via stdin/stdout (JSON-RPC style) or piped HTTP.
+ *
+ * This module provides a "CLI-as-backend" transport: instead of intercepting
+ * HTTP API calls, OmniRoute spawns the CLI directly and feeds prompts through
+ * its native interface.
+ */
+
+import { spawn, ChildProcess } from "child_process";
+import { EventEmitter } from "events";
+
+export interface AcpSession {
+  /** Unique session ID */
+  id: string;
+  /** Agent ID (e.g., "codex", "claude") */
+  agentId: string;
+  /** Child process handle */
+  process: ChildProcess;
+  /** Whether the process is alive */
+  alive: boolean;
+  /** Accumulated stdout buffer */
+  stdoutBuffer: string;
+  /** Accumulated stderr buffer */
+  stderrBuffer: string;
+  /** Created timestamp */
+  createdAt: Date;
+}
+
+/**
+ * ACP Session Manager
+ *
+ * Manages the lifecycle of CLI agent processes.
+ * Each session represents one running CLI agent instance.
+ */
+export class AcpManager extends EventEmitter {
+  private sessions: Map<string, AcpSession> = new Map();
+
+  /**
+   * Spawn a new CLI agent process.
+   */
+  spawn(
+    agentId: string,
+    binary: string,
+    args: string[] = [],
+    env: Record<string, string> = {}
+  ): AcpSession {
+    const sessionId = `acp-${agentId}-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`;
+
+    const child = spawn(binary, args, {
+      stdio: ["pipe", "pipe", "pipe"],
+      env: { ...process.env, ...env },
+      shell: false,
+    });
+
+    const session: AcpSession = {
+      id: sessionId,
+      agentId,
+      process: child,
+      alive: true,
+      stdoutBuffer: "",
+      stderrBuffer: "",
+      createdAt: new Date(),
+    };
+
+    child.stdout?.on("data", (chunk: Buffer) => {
+      session.stdoutBuffer += chunk.toString();
+      this.emit("stdout", { sessionId, data: chunk.toString() });
+    });
+
+    child.stderr?.on("data", (chunk: Buffer) => {
+      session.stderrBuffer += chunk.toString();
+      this.emit("stderr", { sessionId, data: chunk.toString() });
+    });
+
+    child.on("exit", (code, signal) => {
+      session.alive = false;
+      this.emit("exit", { sessionId, code, signal });
+    });
+
+    child.on("error", (err) => {
+      session.alive = false;
+      this.emit("error", { sessionId, error: err });
+    });
+
+    this.sessions.set(sessionId, session);
+    return session;
+  }
+
+  /**
+   * Send input to a running session's stdin.
+   */
+  sendInput(sessionId: string, input: string): boolean {
+    const session = this.sessions.get(sessionId);
+    if (!session?.alive || !session.process.stdin?.writable) return false;
+
+    session.process.stdin.write(input);
+    return true;
+  }
+
+  /**
+   * Send a prompt to a CLI agent and collect the response.
+   * This is a higher-level method that handles the send/receive cycle.
+   */
+  async sendPrompt(sessionId: string, prompt: string, timeoutMs: number = 120000): Promise<string> {
+    const session = this.sessions.get(sessionId);
+    if (!session?.alive) throw new Error(`Session ${sessionId} is not alive`);
+
+    // Clear buffer before sending
+    session.stdoutBuffer = "";
+
+    // Send prompt
+    this.sendInput(sessionId, prompt + "\n");
+
+    // Wait for response (collect until process goes idle or timeout)
+    return new Promise((resolve, reject) => {
+      const timer = setTimeout(() => {
+        reject(new Error(`ACP timeout after ${timeoutMs}ms`));
+      }, timeoutMs);
+
+      let idleTimer: ReturnType<typeof setTimeout>;
+
+      const onData = ({ sessionId: sid }: { sessionId: string }) => {
+        if (sid !== sessionId) return;
+        // Reset idle timer on new data
+        clearTimeout(idleTimer);
+        idleTimer = setTimeout(() => {
+          clearTimeout(timer);
+          this.removeListener("stdout", onData);
+          this.removeListener("exit", onExit);
+          resolve(session.stdoutBuffer);
+        }, 2000); // 2s idle = response complete
+      };
+
+      const onExit = ({ sessionId: sid }: { sessionId: string }) => {
+        if (sid !== sessionId) return;
+        clearTimeout(timer);
+        clearTimeout(idleTimer);
+        this.removeListener("stdout", onData);
+        this.removeListener("exit", onExit);
+        resolve(session.stdoutBuffer);
+      };
+
+      this.on("stdout", onData);
+      this.on("exit", onExit);
+    });
+  }
+
+  /**
+   * Kill a session and clean up.
+   */
+  kill(sessionId: string): boolean {
+    const session = this.sessions.get(sessionId);
+    if (!session) return false;
+
+    if (session.alive) {
+      session.process.kill("SIGTERM");
+      // Force kill after 5s
+      setTimeout(() => {
+        if (session.alive) {
+          session.process.kill("SIGKILL");
+        }
+      }, 5000);
+    }
+
+    this.sessions.delete(sessionId);
+    return true;
+  }
+
+  /**
+   * Get all active sessions.
+   */
+  getActiveSessions(): AcpSession[] {
+    return Array.from(this.sessions.values()).filter((s) => s.alive);
+  }
+
+  /**
+   * Get a specific session.
+   */
+  getSession(sessionId: string): AcpSession | undefined {
+    return this.sessions.get(sessionId);
+  }
+
+  /**
+   * Kill all sessions.
+   */
+  killAll(): void {
+    for (const [id] of this.sessions) {
+      this.kill(id);
+    }
+  }
+}
+
+// Singleton manager instance
+export const acpManager = new AcpManager();
@@ -0,0 +1,126 @@
+/**
+ * ACP (Agent Client Protocol) — CLI Agent Registry
+ *
+ * Discovers installed CLI tools on the system by checking standard paths
+ * and running version commands. Used to offer ACP transport as an alternative
+ * to the HTTP proxy method.
+ *
+ * Reference: https://github.com/iOfficeAI/AionUi (auto-detects CLI agents)
+ */
+
+import { execSync } from "child_process";
+
+export interface CliAgentInfo {
+  /** Agent identifier (e.g., "codex", "claude", "goose") */
+  id: string;
+  /** Display name */
+  name: string;
+  /** Binary name to spawn */
+  binary: string;
+  /** Version detection command */
+  versionCommand: string;
+  /** Detected version (null if not installed) */
+  version: string | null;
+  /** Whether the agent is installed and available */
+  installed: boolean;
+  /** Provider ID that this agent maps to in OmniRoute */
+  providerAlias: string;
+  /** Arguments to pass when spawning for ACP */
+  spawnArgs: string[];
+  /** Protocol used for communication */
+  protocol: "stdio" | "http";
+}
+
+/**
+ * Registry of known CLI agents that support ACP or similar protocols.
+ */
+const AGENT_DEFINITIONS: Omit<CliAgentInfo, "version" | "installed">[] = [
+  {
+    id: "codex",
+    name: "OpenAI Codex CLI",
+    binary: "codex",
+    versionCommand: "codex --version",
+    providerAlias: "codex",
+    spawnArgs: ["--quiet"],
+    protocol: "stdio",
+  },
+  {
+    id: "claude",
+    name: "Claude Code CLI",
+    binary: "claude",
+    versionCommand: "claude --version",
+    providerAlias: "claude",
+    spawnArgs: ["--print", "--output-format", "json"],
+    protocol: "stdio",
+  },
+  {
+    id: "goose",
+    name: "Goose CLI",
+    binary: "goose",
+    versionCommand: "goose --version",
+    providerAlias: "goose",
+    spawnArgs: [],
+    protocol: "stdio",
+  },
+  {
+    id: "gemini-cli",
+    name: "Gemini CLI",
+    binary: "gemini",
+    versionCommand: "gemini --version",
+    providerAlias: "gemini-cli",
+    spawnArgs: [],
+    protocol: "stdio",
+  },
+  {
+    id: "openclaw",
+    name: "OpenClaw",
+    binary: "openclaw",
+    versionCommand: "openclaw --version",
+    providerAlias: "openclaw",
+    spawnArgs: [],
+    protocol: "stdio",
+  },
+];
+
+/**
+ * Detect installed CLI agents on the system.
+ * Runs version commands to verify availability.
+ */
+export function detectInstalledAgents(): CliAgentInfo[] {
+  return AGENT_DEFINITIONS.map((def) => {
+    let version: string | null = null;
+    let installed = false;
+
+    try {
+      const output = execSync(def.versionCommand, {
+        timeout: 5000,
+        encoding: "utf-8",
+        stdio: ["pipe", "pipe", "pipe"],
+      }).trim();
+
+      // Extract version number from output
+      const versionMatch = output.match(/(\d+\.\d+\.\d+(?:-\w+)?)/);
+      version = versionMatch ? versionMatch[1] : output.split("\n")[0];
+      installed = true;
+    } catch {
+      // Not installed or not runnable
+    }
+
+    return { ...def, version, installed };
+  });
+}
+
+/**
+ * Get a specific agent by ID.
+ */
+export function getAgentById(id: string): CliAgentInfo | undefined {
+  const agents = detectInstalledAgents();
+  return agents.find((a) => a.id === id);
+}
+
+/**
+ * Get agents that are installed and available for ACP.
+ */
+export function getAvailableAgents(): CliAgentInfo[] {
+  return detectInstalledAgents().filter((a) => a.installed);
+}
@@ -79,6 +79,7 @@ const SCHEMA_SQL = `
    token_type TEXT,
    consecutive_use_count INTEGER DEFAULT 0,
    rate_limit_protection INTEGER DEFAULT 0,
+    last_used_at TEXT,
    created_at TEXT NOT NULL,
    updated_at TEXT NOT NULL
  );
@@ -311,6 +312,10 @@ function ensureProviderConnectionsColumns(db: SqliteDatabase) {
      );
      console.log("[DB] Added provider_connections.rate_limit_protection column");
    }
+    if (!columnNames.has("last_used_at")) {
+      db.exec("ALTER TABLE provider_connections ADD COLUMN last_used_at TEXT");
+      console.log("[DB] Added provider_connections.last_used_at column");
+    }
  } catch (error: unknown) {
    const message = error instanceof Error ? error.message : String(error);
    console.warn("[DB] Failed to verify provider_connections schema:", message);
@@ -483,7 +488,7 @@ function migrateFromJson(db: SqliteDatabase, jsonPath: string) {
          rate_limited_until, health_check_interval, last_health_check_at,
          last_tested, api_key, id_token, provider_specific_data,
          expires_in, display_name, global_priority, default_model,
-          token_type, consecutive_use_count, rate_limit_protection, created_at, updated_at
+          token_type, consecutive_use_count, rate_limit_protection, last_used_at, created_at, updated_at
        ) VALUES (
          @id, @provider, @authType, @name, @email, @priority, @isActive,
          @accessToken, @refreshToken, @expiresAt, @tokenExpiresAt,
@@ -492,7 +497,7 @@ function migrateFromJson(db: SqliteDatabase, jsonPath: string) {
          @rateLimitedUntil, @healthCheckInterval, @lastHealthCheckAt,
          @lastTested, @apiKey, @idToken, @providerSpecificData,
          @expiresIn, @displayName, @globalPriority, @defaultModel,
-          @tokenType, @consecutiveUseCount, @rateLimitProtection, @createdAt, @updatedAt
+          @tokenType, @consecutiveUseCount, @rateLimitProtection, @lastUsedAt, @createdAt, @updatedAt
        )
      `);

@@ -533,6 +538,7 @@ function migrateFromJson(db: SqliteDatabase, jsonPath: string) {
          defaultModel: conn.defaultModel || null,
          tokenType: conn.tokenType || null,
          consecutiveUseCount: conn.consecutiveUseCount || 0,
+          lastUsedAt: conn.lastUsedAt || null,
          rateLimitProtection:
            conn.rateLimitProtection === true || conn.rateLimitProtection === 1 ? 1 : 0,
          createdAt: conn.createdAt || new Date().toISOString(),
@@ -115,7 +115,14 @@ export async function getAllCustomModels() {
  return result;
 }

-export async function addCustomModel(providerId, modelId, modelName, source = "manual") {
+export async function addCustomModel(
+  providerId: string,
+  modelId: string,
+  modelName?: string,
+  source = "manual",
+  apiFormat: "chat-completions" | "responses" = "chat-completions",
+  supportedEndpoints: string[] = ["chat"]
+) {
  const db = getDbInstance();
  const row = db
    .prepare("SELECT value FROM key_value WHERE namespace = 'customModels' AND key = ?")
@@ -126,7 +133,13 @@ export async function addCustomModel(providerId, modelId, modelName, source = "m
  const exists = models.find((m) => m.id === modelId);
  if (exists) return exists;

-  const model = { id: modelId, name: modelName || modelId, source };
+  const model = {
+    id: modelId,
+    name: modelName || modelId,
+    source,
+    apiFormat,
+    supportedEndpoints,
+  };
  models.push(model);
  db.prepare(
    "INSERT OR REPLACE INTO key_value (namespace, key, value) VALUES ('customModels', ?, ?)"
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
diegosouzapw	cbd60c853e	feat(release): v2.0.10 — CLI fingerprint UI toggle Build Electron Desktop App / Validate version (push) Failing after 27s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-07 10:57:16 -03:00
diegosouzapw	2d8091340f	feat: CLI fingerprint UI toggle + fix playground i18n (#223 ) - Add CLI Fingerprint Matching card in Settings > Security tab - Per-provider toggle chips (codex, claude, github, antigravity) - Emerald-themed UI with fingerprint icon and active count - Settings synced to runtime cache via API - Fix playground i18n missing from 29 non-English sidebar translations - Add cliCompatProviders to settings schema (Zod validation) - Add setCliCompatProviders/getCliCompatProviders to cliFingerprints.ts - Add i18n keys for CLI fingerprint settings (en + pt-BR)	2026-03-07 10:56:58 -03:00
diegosouzapw	2025c16c82	fix: deploy workflow uses node app/server.js for pm2	2026-03-07 10:40:00 -03:00
diegosouzapw	811fb7f9b2	docs: fix deploy workflow to use npm-only approach Old workflow used git clone at /opt/omniroute-app but recent releases used npm install -g, causing version mismatch. PM2 now runs from /usr/lib/node_modules/omniroute directly. Removed all references to the stale local directory.	2026-03-07 10:38:49 -03:00
diegosouzapw	a2bd32e76c	feat(release): v2.0.9 — playground, CLI fingerprints, ACP Build Electron Desktop App / Validate version (push) Failing after 27s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-07 10:26:34 -03:00
Diego Rodrigues de Sa e Souza	89c07f4e8a	Merge pull request #240 from diegosouzapw/feat/all-features-234-223-235 feat: playground, CLI fingerprints, ACP support (#234, #223, #235)	2026-03-07 10:25:28 -03:00
diegosouzapw	ac89069671	feat: playground, CLI fingerprints, ACP support (#234 , #223 , #235 ) - Add model playground page to dashboard (provider/model/endpoint selectors, Monaco editor, streaming, abort) - Add CLI fingerprint matching (per-provider header/body ordering to match native CLI binaries) - Add ACP module (CLI agent discovery, process spawner, session manager, API endpoint) - Add playground to sidebar debug section with i18n support - Close #192 and #200 as stale (v1.8.1, no repro info)	2026-03-07 10:24:43 -03:00
diegosouzapw	7cb420d8e6	feat(release): v2.0.8 — custom image model handler resolution Build Electron Desktop App / Validate version (push) Failing after 26s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-07 10:05:20 -03:00
Diego Rodrigues de Sa e Souza	d3919d441f	Merge pull request #239 from diegosouzapw/fix/issue-238-image-handler fix: pass resolved provider to image handler for custom models (#238)	2026-03-07 10:04:24 -03:00
diegosouzapw	4b5824babc	fix: pass resolved provider to image handler for custom models (#238 )	2026-03-07 10:03:48 -03:00
diegosouzapw	fb87df14fd	feat(release): v2.0.7 — custom image model routing + Codex OAuth workspace isolation Build Electron Desktop App / Validate version (push) Failing after 34s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-07 06:58:07 -03:00
Diego Rodrigues de Sa e Souza	da9e4e929b	Merge pull request #237 from diegosouzapw/fix/issue-232-236-image-oauth fix: custom image model routing + Codex OAuth workspace isolation (#232, #236)	2026-03-07 06:56:49 -03:00
diegosouzapw	10b23b15ae	fix: custom image model routing + Codex OAuth workspace isolation (#232 , #236 )	2026-03-07 06:56:09 -03:00
diegosouzapw	30fba39b35	feat(release): v2.0.6 — custom model apiFormat routing fix Build Electron Desktop App / Validate version (push) Failing after 33s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-07 01:36:21 -03:00
Diego Rodrigues de Sa e Souza	5a75ff67c9	Merge pull request #233 from diegosouzapw/fix/issue-204-apiformat-routing fix: wire apiFormat from custom model DB into routing layer (#204)	2026-03-07 01:35:30 -03:00
diegosouzapw	358828b617	fix: wire apiFormat from custom model DB into routing layer (#204 )	2026-03-07 01:26:59 -03:00
diegosouzapw	e080c4a16a	feat(release): v2.0.5 — fix Chat→Responses reasoning IDs, electron auto-update, dependency bumps Build Electron Desktop App / Validate version (push) Failing after 31s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-06 18:51:24 -03:00
Diego Rodrigues de Sa e Souza	04b7e38baf	Merge pull request #221 from benzntech/feat/electron-auto-update feat(electron): add auto-update functionality with electron-updater	2026-03-06 18:49:54 -03:00
Diego Rodrigues de Sa e Souza	7ee23fbe19	Merge pull request #230 from diegosouzapw/dependabot/npm_and_yarn/express-rate-limit-8.3.0 deps: bump express-rate-limit from 8.2.1 to 8.3.0	2026-03-06 18:49:51 -03:00
Diego Rodrigues de Sa e Souza	c49bdb4ebb	Merge pull request #229 from diegosouzapw/dependabot/github_actions/docker/build-push-action-7 chore(deps): bump docker/build-push-action from 6 to 7	2026-03-06 18:49:48 -03:00
Diego Rodrigues de Sa e Souza	0f7efed8d5	Merge pull request #228 from diegosouzapw/dependabot/github_actions/actions/upload-artifact-7 chore(deps): bump actions/upload-artifact from 4 to 7	2026-03-06 18:49:46 -03:00
Diego Rodrigues de Sa e Souza	d07bc6dcf3	Merge pull request #227 from diegosouzapw/dependabot/github_actions/docker/login-action-4 chore(deps): bump docker/login-action from 3 to 4	2026-03-06 18:49:43 -03:00
Diego Rodrigues de Sa e Souza	d607d46fa3	Merge pull request #226 from diegosouzapw/dependabot/github_actions/actions/download-artifact-8 chore(deps): bump actions/download-artifact from 4 to 8	2026-03-06 18:49:40 -03:00
Diego Rodrigues de Sa e Souza	2225dd14aa	Merge pull request #225 from diegosouzapw/dependabot/github_actions/actions/cache-5 chore(deps): bump actions/cache from 4 to 5	2026-03-06 18:49:37 -03:00
Diego Rodrigues de Sa e Souza	f6c0e7bbbe	Merge pull request #222 from benzntech/fix/electron-release-duplicate-asset fix(ci): remove duplicate OmniRoute.exe entry in electron release workflow	2026-03-06 18:49:28 -03:00
Diego Rodrigues de Sa e Souza	c4675c5219	Merge pull request #231 from diegosouzapw/fix/issue-224-reasoning-ids fix: omit synthesized reasoning items in Chat→Responses translation (#224)	2026-03-06 18:49:25 -03:00
diegosouzapw	2d977a3c4d	fix: omit synthesized reasoning items in Chat→Responses translation (#224 )	2026-03-06 18:48:34 -03:00
dependabot[bot]	9405918258	deps: bump express-rate-limit from 8.2.1 to 8.3.0 Bumps [express-rate-limit](https://github.com/express-rate-limit/express-rate-limit) from 8.2.1 to 8.3.0. - [Release notes](https://github.com/express-rate-limit/express-rate-limit/releases) - [Commits](https://github.com/express-rate-limit/express-rate-limit/compare/v8.2.1...v8.3.0) --- updated-dependencies: - dependency-name: express-rate-limit dependency-version: 8.3.0 dependency-type: indirect ... Signed-off-by: dependabot[bot] <support@github.com>	2026-03-06 18:46:36 +00:00
dependabot[bot]	a69d7dd4b5	chore(deps): bump docker/build-push-action from 6 to 7 Bumps [docker/build-push-action](https://github.com/docker/build-push-action) from 6 to 7. - [Release notes](https://github.com/docker/build-push-action/releases) - [Commits](https://github.com/docker/build-push-action/compare/v6...v7) --- updated-dependencies: - dependency-name: docker/build-push-action dependency-version: '7' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>	2026-03-06 18:27:03 +00:00
dependabot[bot]	428e6cb53f	chore(deps): bump actions/upload-artifact from 4 to 7 Bumps [actions/upload-artifact](https://github.com/actions/upload-artifact) from 4 to 7. - [Release notes](https://github.com/actions/upload-artifact/releases) - [Commits](https://github.com/actions/upload-artifact/compare/v4...v7) --- updated-dependencies: - dependency-name: actions/upload-artifact dependency-version: '7' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>	2026-03-06 18:26:59 +00:00
dependabot[bot]	c9a2955d28	chore(deps): bump docker/login-action from 3 to 4 Bumps [docker/login-action](https://github.com/docker/login-action) from 3 to 4. - [Release notes](https://github.com/docker/login-action/releases) - [Commits](https://github.com/docker/login-action/compare/v3...v4) --- updated-dependencies: - dependency-name: docker/login-action dependency-version: '4' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>	2026-03-06 18:26:54 +00:00
dependabot[bot]	7aefcd3437	chore(deps): bump actions/download-artifact from 4 to 8 Bumps [actions/download-artifact](https://github.com/actions/download-artifact) from 4 to 8. - [Release notes](https://github.com/actions/download-artifact/releases) - [Commits](https://github.com/actions/download-artifact/compare/v4...v8) --- updated-dependencies: - dependency-name: actions/download-artifact dependency-version: '8' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>	2026-03-06 18:26:51 +00:00
dependabot[bot]	79f4f79c46	chore(deps): bump actions/cache from 4 to 5 Bumps [actions/cache](https://github.com/actions/cache) from 4 to 5. - [Release notes](https://github.com/actions/cache/releases) - [Changelog](https://github.com/actions/cache/blob/main/RELEASES.md) - [Commits](https://github.com/actions/cache/compare/v4...v5) --- updated-dependencies: - dependency-name: actions/cache dependency-version: '5' dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>	2026-03-06 18:26:46 +00:00
benzntech	c11c275678	fix(electron): address auto-updater review issues - Remove unused dialog import - Stop Next.js server before quitAndInstall() to prevent data loss - Propagate errors from checkForUpdates/downloadUpdate to IPC handlers so renderer can distinguish success from failure - Remove meaningless return value from install-update handler	2026-03-06 19:22:41 +05:30
benzntech	bbcd1d3a08	fix(ci): remove duplicate OmniRoute.exe entry in electron release workflow Duplicate release-assets/OmniRoute.exe glob caused softprops/action-gh-release to attempt a second upload of the same asset, triggering a 404 Not Found error on the GitHub release asset update API. The file is already covered by the *.exe glob pattern above it.	2026-03-06 19:18:41 +05:30
benzntech	3342d5b931	feat(electron): add auto-update functionality with electron-updater	2026-03-06 18:54:00 +05:30
diegosouzapw	f96ee44213	feat(release): v2.0.4 — round-robin lastUsedAt persistence, zod standalone build fix Build Electron Desktop App / Validate version (push) Failing after 31s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-05 23:24:56 -03:00
Diego Rodrigues de Sa e Souza	bc53fe0cd9	Merge pull request #219 from diegosouzapw/fix/issue-218-round-robin-lastUsedAt fix: persist lastUsedAt for round-robin + zod in standalone build (#218, #217)	2026-03-05 23:24:13 -03:00
diegosouzapw	97a67b5d3e	fix: persist lastUsedAt in provider_connections schema for round-robin (#218 ) - Add last_used_at column to provider_connections CREATE TABLE schema - Add ensureProviderConnectionsColumns migration for existing databases - Add last_used_at to INSERT and UPDATE SQL in providers.ts - Add last_used_at to JSON migration INSERT in core.ts - Add zod to serverExternalPackages in next.config.mjs (#217) Fixes #218: Round-robin routing strategy now correctly persists the lastUsedAt timestamp, allowing rotation between accounts. Fixes #217: zod module is now properly included in standalone/Docker builds by declaring it as a server external package.	2026-03-05 23:22:10 -03:00
diegosouzapw	1ffa58be76	feat(release): v2.0.3 — deferred tools cache_control fix, quota system hardening Build Electron Desktop App / Validate version (push) Failing after 29s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-05 21:57:04 -03:00
Diego Rodrigues de Sa e Souza	a5cf51c0b9	Merge pull request #214 from DavyMassoneto/fix/claude-oauth-usage-endpoint fix: harden quota system — code review fixes + build fix	2026-03-05 21:55:23 -03:00
Diego Rodrigues de Sa e Souza	3d38cbf70f	Merge pull request #216 from DavyMassoneto/fix/defer-loading-cache-control-conflict fix: skip cache_control on deferred tools + remove stale schemas.js	2026-03-05 21:55:14 -03:00
DavyMassoneto	196a4e037c	fix: skip cache_control on deferred tools + remove stale schemas.js - Skip tools with defer_loading=true when assigning cache_control (Anthropic API rejects the combination with 400) - Delete stale schemas.js that shadowed the .ts source, causing missing cloudSyncActionSchema export Fixes #215	2026-03-05 20:19:58 -03:00
diegosouzapw	6a0760a2c5	chore: bump to v2.0.2 (v2.0.1 was claimed on npm) Build Electron Desktop App / Validate version (push) Failing after 40s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-05 20:11:15 -03:00
diegosouzapw	b0819404c7	feat(release): v2.0.1 — endpoint-aware models, 3 bug fixes (#212 , #213 , #200 ), 3 features (#204 , #205 , #206 ) Build Electron Desktop App / Validate version (push) Failing after 35s Details Build Electron Desktop App / Build Electron (macos-arm64) (push) Has been skipped Details Build Electron Desktop App / Build Electron (linux) (push) Has been skipped Details Build Electron Desktop App / Build Electron (macos-intel) (push) Has been skipped Details Build Electron Desktop App / Build Electron (windows) (push) Has been skipped Details Build Electron Desktop App / Create Release (push) Has been skipped Details	2026-03-05 20:04:41 -03:00
DavyMassoneto	bfe495931f	fix(claude): correct utilization semantics, harden quota cache, fix premature model unavailability - Fix inverted Claude OAuth utilization (remaining, not used) - Add hasUtilization() guard to prevent false exhaustion from empty responses - Centralize anthropic-version into CLAUDE_CONFIG.apiVersion - Add parseDate() for safe date validation in quota cache - Batch background refresh with MAX_CONCURRENT_REFRESHES=5 - Move setModelUnavailable to after all accounts exhausted, not first 429 - Extract safePercentage() to shared utils (dedup) - Use isRecord() type guard in usage API route - Exclude binary files from Tailwind v4 source scanning	2026-03-05 19:39:59 -03:00
DavyMassoneto	11bcdd810a	feat: quota-aware account selection + fix premature model unavailability - Move setModelUnavailable from per-account loop to all-accounts-exhausted path - Clear model unavailability on successful fallback - Add in-memory quota cache with background refresh (5min active, 20min exhausted) - Integrate quota cache in account selection to skip exhausted accounts - Mark accounts as exhausted from 429 when no cached quota data exists - Populate quota cache from dashboard usage endpoint	2026-03-05 18:49:56 -03:00
diegosouzapw	228ebf436e	feat: endpoint-aware model management + fix 3 bugs (#212 , #213 , #200 ) Bug Fixes: - #212: Auto-generate API_KEY_SECRET at startup (like JWT_SECRET) - #213: Circuit breaker now scoped per-model instead of per-provider - #200: Connectivity fallback for custom providers (Ollama, LM Studio) Features: - #204: API Format selector (Chat Completions / Responses API) for custom models - #205: Combo endpoint field (chat / embeddings / images) in schema - #206: Supported Endpoints checkboxes (chat, embeddings, images, audio) - Custom models with endpoint tags appear in /v1/embeddings and /v1/images/generations - Model catalog includes api_format, type, and supported_endpoints metadata - Provider detail page shows badges for non-default endpoint configurations Files changed: instrumentation.ts, combo.ts, validation.ts, models.ts, schemas.ts, provider-models/route.ts, providers/[id]/page.tsx, catalog.ts, embeddings/route.ts, images/generations/route.ts	2026-03-05 18:49:07 -03:00