feat(beszel): expose Beszel systems, alerts, and stats via MCP

Register beszelRouter in the MCP server with .meta({ mcp: enabled })
on getSystems, getAlerts, and getSystemStats procedures. AI assistants
can now query Beszel-monitored systems, view alerts, and fetch
historical metrics through the MCP protocol.

Ref #5968

apps/docs/docs/management/mcp.mdx

@@ -130,6 +130,7 @@ Homarr exposes 50+ tools organized by namespace. Each tool is either a **query**
 | ------------------ | -------------------------------------------------------- | ---------- |
 | `app`              | List all apps, get app by ID, create/update/delete app   | GET + POST |
 | `apiKeys`          | List API keys                                            | GET        |
+| `beszel`           | Systems status, alerts, historical system metrics        | GET        |
 | `board`            | List boards, get board by ID, create/update/delete board | GET + POST |
 | `docker`           | List containers, inspect container, start/stop/restart   | GET + POST |
 | `icon`             | Search icons                                             | GET        |
@@ -162,3 +163,4 @@ The tools table on the MCP tab is dynamically generated - it always reflects the
 - **"Create a new app pointing to my Grafana instance"** - AI creates an app with the correct icon and URL
 - **"How many users are on my Homarr instance?"** - AI lists all users (admin only)
 - **"Toggle Pi-hole blocking for 5 minutes"** - AI interacts with DNS controls
+- **"How are my servers doing?"** - AI fetches Beszel system stats including CPU, memory, disk, and alerts

apps/nextjs/src/app/api/mcp/[transport]/route.ts

@@ -148,6 +148,8 @@ If \`hasUseAccess\` is false, the API key owner lacks permission for that integr
 
 **"Check server health"** → Use \`healthMonitoring_getSystemHealthStatus\` for NAS/server metrics or \`healthMonitoring_getClusterHealthStatus\` for Proxmox clusters.
 
+**"Check Beszel systems"** → Use \`beszel_getSystems\` to list all monitored systems with CPU/memory/disk/network status. Use \`beszel_getAlerts\` for active alerts and history. Use \`beszel_getSystemStats\` for historical metrics of a specific system (requires systemId from \`beszel_getSystems\`).
+
 **"Manage Docker containers"** → Use \`docker_getContainers\` to list containers with status/CPU/memory, then \`docker_startAll\`/\`docker_stopAll\`/\`docker_restartAll\` with container IDs.
 
 **"Control smart home"** → Use \`smartHome_entityState\` to check a Home Assistant entity, \`smartHome_switchEntity\` to toggle it, or \`smartHome_executeAutomation\` to trigger an automation.

packages/api/src/mcp.ts

@@ -16,6 +16,7 @@ import { healthMonitoringRouter } from "./router/widgets/health-monitoring";
 import { mediaRequestsRouter } from "./router/widgets/media-requests";
 import { mediaServerRouter } from "./router/widgets/media-server";
 import { smartHomeRouter } from "./router/widgets/smart-home";
+import { beszelRouter } from "./router/widgets/beszel";
 
 export const mcpRouter = createTRPCRouter({
   app: appRouterForApps,
@@ -26,6 +27,7 @@ export const mcpRouter = createTRPCRouter({
   info: infoRouter,
   integration: integrationRouter,
   invite: inviteRouter,
+  beszel: beszelRouter,
   calendar: calendarRouter,
   dnsHole: dnsHoleRouter,
   downloads: downloadsRouter,

packages/api/src/router/widgets/beszel.ts

@@ -15,6 +15,13 @@ import { createTRPCRouter, publicProcedure } from "../../trpc";
 
 export const beszelRouter = createTRPCRouter({
   getSystems: publicProcedure
+    .meta({
+      mcp: {
+        enabled: true,
+        description:
+          "Get all Beszel-monitored systems with CPU, memory, disk, GPU, network, temperature, and status. REQUIRED: integrationIds (array of Beszel integration IDs from integration_all)",
+      },
+    })
     .concat(createManyIntegrationMiddleware("query", "beszel", "mock"))
     .query(async ({ ctx }) => {
       const results = await Promise.all(
@@ -60,6 +67,13 @@ export const beszelRouter = createTRPCRouter({
     }),
 
   getAlerts: publicProcedure
+    .meta({
+      mcp: {
+        enabled: true,
+        description:
+          "Get Beszel alerts and optional alert history for all monitored systems. REQUIRED: integrationIds (array of Beszel integration IDs from integration_all). OPTIONAL: includeHistory (default true), maxHistoryItems (default 10)",
+      },
+    })
     .concat(createManyIntegrationMiddleware("query", "beszel", "mock"))
     .input(
       z.object({
@@ -129,6 +143,13 @@ export const beszelRouter = createTRPCRouter({
     }),
 
   getSystemStats: publicProcedure
+    .meta({
+      mcp: {
+        enabled: true,
+        description:
+          "Get historical Beszel system metrics (CPU, memory, disk, network, temperature) and optional Docker container stats. REQUIRED: integrationIds (array of Beszel integration IDs from integration_all), systemId (from beszel_getSystems). OPTIONAL: timePeriod (1h/12h/24h/1w/30d, default 1h), includeDocker (default true)",
+      },
+    })
     .concat(createManyIntegrationMiddleware("query", "beszel", "mock"))
     .input(
       z.object({