feat: add LiteLLM gateway scaffold and dashboard health probe

Introduce a localhost-only LiteLLM service template and expose a protected dashboard health endpoint for operational visibility, while adding cc-connect mobile bridge templates with conservative defaults.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

dashboard/backend/config.py

@@ -44,6 +44,9 @@ CORS_ORIGINS = os.environ.get("CORS_ORIGINS", "https://nas.jimmygan.com,https://
 # Chat Summary
 CHAT_SUMMARY_DB = os.environ.get("CHAT_SUMMARY_DB", "/app/data/chat-summarizer/chat_summary.db")
 
+# LiteLLM
+LITELLM_URL = os.environ.get("LITELLM_URL", "http://127.0.0.1:4005")
+
 # OpenClaw
 OPENCLAW_GATEWAY_TOKEN = os.environ.get("OPENCLAW_GATEWAY_TOKEN", "")
 

dashboard/backend/main.py

@@ -6,7 +6,7 @@ from fastapi.responses import JSONResponse
 from slowapi import Limiter
 from slowapi.util import get_remote_address
 from slowapi.errors import RateLimitExceeded
-from routers import docker_router, gitea, files, terminal, system, auth, chat_summary, security, passkey, totp
+from routers import docker_router, gitea, files, terminal, system, auth, chat_summary, security, passkey, totp, litellm
 import auth as auth_module
 from rbac import require_page, require_write
 
@@ -179,6 +179,8 @@ app.include_router(files.router, prefix="/api/files",
     dependencies=[Depends(_inject_user), Depends(require_page("files")), Depends(require_write())])
 app.include_router(system.router, prefix="/api/system",
     dependencies=[Depends(_inject_user), Depends(require_page("dashboard"))])
+app.include_router(litellm.router, prefix="/api/litellm",
+    dependencies=[Depends(_inject_user), Depends(require_page("dashboard"))])
 app.include_router(chat_summary.router, prefix="/api/chat-summary",
     dependencies=[Depends(_inject_user), Depends(require_page("chat-digest"))])
 app.include_router(security.router, prefix="/api/security",

dashboard/backend/routers/litellm.py

@@ -0,0 +1,40 @@
+import time
+import httpx
+from fastapi import APIRouter
+from config import LITELLM_URL
+
+router = APIRouter()
+
+
+@router.get("/health")
+async def litellm_health():
+    timeout = httpx.Timeout(2.0)
+    checks = ("/health", "/v1/models")
+
+    last_error = None
+    async with httpx.AsyncClient(timeout=timeout) as client:
+        for path in checks:
+            url = f"{LITELLM_URL.rstrip('/')}{path}"
+            started = time.perf_counter()
+            try:
+                response = await client.get(url)
+                latency_ms = round((time.perf_counter() - started) * 1000, 1)
+                if response.is_success:
+                    return {
+                        "ok": True,
+                        "status": "up",
+                        "latency_ms": latency_ms,
+                        "endpoint": path,
+                        "http_status": response.status_code,
+                    }
+                last_error = f"{path} returned {response.status_code}"
+            except Exception as exc:
+                last_error = str(exc)
+
+    return {
+        "ok": False,
+        "status": "down",
+        "latency_ms": None,
+        "endpoint": None,
+        "error": last_error or "health check failed",
+    }