feat(metrics): add Prometheus /metrics endpoint

Zero-dependency Prometheus text exposition format. Exposes monitor
up/down, latency, status code, check timestamps, pause state,
SSL cert expiry, and check counters — all from in-memory state.

internal/metrics/prometheus.go

@@ -0,0 +1,99 @@
+package metrics
+
+import (
+	"fmt"
+	"go-upkeep/internal/models"
+	"go-upkeep/internal/monitor"
+	"net/http"
+	"sort"
+	"strings"
+)
+
+func Handler(eng *monitor.Engine) http.HandlerFunc {
+	return func(w http.ResponseWriter, r *http.Request) {
+		sites := eng.GetAllSites()
+		sort.Slice(sites, func(i, j int) bool { return sites[i].ID < sites[j].ID })
+
+		var b strings.Builder
+
+		writeHelp(&b, "upkeep_monitor_up", "gauge", "Whether the monitor is up (1) or down (0).")
+		for _, s := range sites {
+			val := 0
+			if s.Status == "UP" {
+				val = 1
+			}
+			writeGauge(&b, "upkeep_monitor_up", labels(s), float64(val))
+		}
+
+		writeHelp(&b, "upkeep_monitor_latency_seconds", "gauge", "Last check latency in seconds.")
+		for _, s := range sites {
+			writeGauge(&b, "upkeep_monitor_latency_seconds", labels(s), s.Latency.Seconds())
+		}
+
+		writeHelp(&b, "upkeep_monitor_status_code", "gauge", "HTTP response status code of the last check.")
+		for _, s := range sites {
+			if s.Type != "http" {
+				continue
+			}
+			writeGauge(&b, "upkeep_monitor_status_code", labels(s), float64(s.StatusCode))
+		}
+
+		writeHelp(&b, "upkeep_monitor_check_timestamp_seconds", "gauge", "Unix timestamp of the last check.")
+		for _, s := range sites {
+			if s.LastCheck.IsZero() {
+				continue
+			}
+			writeGauge(&b, "upkeep_monitor_check_timestamp_seconds", labels(s), float64(s.LastCheck.Unix()))
+		}
+
+		writeHelp(&b, "upkeep_monitor_paused", "gauge", "Whether the monitor is paused (1) or active (0).")
+		for _, s := range sites {
+			val := 0
+			if s.Paused {
+				val = 1
+			}
+			writeGauge(&b, "upkeep_monitor_paused", labels(s), float64(val))
+		}
+
+		writeHelp(&b, "upkeep_monitor_cert_expiry_timestamp_seconds", "gauge", "Unix timestamp when the SSL certificate expires.")
+		for _, s := range sites {
+			if !s.HasSSL || s.CertExpiry.IsZero() {
+				continue
+			}
+			writeGauge(&b, "upkeep_monitor_cert_expiry_timestamp_seconds", labels(s), float64(s.CertExpiry.Unix()))
+		}
+
+		writeHelp(&b, "upkeep_monitor_checks_total", "counter", "Total number of checks performed.")
+		writeHelp(&b, "upkeep_monitor_checks_up_total", "counter", "Total number of successful checks.")
+		for _, s := range sites {
+			h, ok := eng.GetHistory(s.ID)
+			if !ok {
+				continue
+			}
+			writeGauge(&b, "upkeep_monitor_checks_total", labels(s), float64(h.TotalChecks))
+			writeGauge(&b, "upkeep_monitor_checks_up_total", labels(s), float64(h.UpChecks))
+		}
+
+		w.Header().Set("Content-Type", "text/plain; version=0.0.4; charset=utf-8")
+		w.Write([]byte(b.String()))
+	}
+}
+
+func labels(s models.Site) string {
+	return fmt.Sprintf(`id="%d",name="%s",type="%s"`, s.ID, escapeLabelValue(s.Name), s.Type)
+}
+
+func escapeLabelValue(s string) string {
+	s = strings.ReplaceAll(s, `\`, `\\`)
+	s = strings.ReplaceAll(s, `"`, `\"`)
+	s = strings.ReplaceAll(s, "\n", `\n`)
+	return s
+}
+
+func writeHelp(b *strings.Builder, name, typ, help string) {
+	fmt.Fprintf(b, "# HELP %s %s\n# TYPE %s %s\n", name, help, name, typ)
+}
+
+func writeGauge(b *strings.Builder, name, labels string, val float64) {
+	fmt.Fprintf(b, "%s{%s} %g\n", name, labels, val)
+}

internal/metrics/prometheus_test.go

@@ -0,0 +1,96 @@
+package metrics
+
+import (
+	"context"
+	"go-upkeep/internal/models"
+	"go-upkeep/internal/monitor"
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"testing"
+	"time"
+)
+
+type mockStore struct {
+	sites []models.Site
+}
+
+func (m *mockStore) Init() error                                              { return nil }
+func (m *mockStore) GetSites() ([]models.Site, error)                         { return m.sites, nil }
+func (m *mockStore) AddSite(models.Site) error                                { return nil }
+func (m *mockStore) UpdateSite(models.Site) error                             { return nil }
+func (m *mockStore) UpdateSitePaused(int, bool) error                         { return nil }
+func (m *mockStore) DeleteSite(int) error                                     { return nil }
+func (m *mockStore) GetAllAlerts() ([]models.AlertConfig, error)              { return nil, nil }
+func (m *mockStore) GetAlert(int) (models.AlertConfig, error)                 { return models.AlertConfig{}, nil }
+func (m *mockStore) AddAlert(string, string, map[string]string) error         { return nil }
+func (m *mockStore) UpdateAlert(int, string, string, map[string]string) error { return nil }
+func (m *mockStore) DeleteAlert(int) error                                    { return nil }
+func (m *mockStore) GetAllUsers() ([]models.User, error)                      { return nil, nil }
+func (m *mockStore) AddUser(string, string, string) error                     { return nil }
+func (m *mockStore) UpdateUser(int, string, string, string) error             { return nil }
+func (m *mockStore) DeleteUser(int) error                                     { return nil }
+func (m *mockStore) SaveCheck(int, int64, bool) error                         { return nil }
+func (m *mockStore) LoadAllHistory(int) (map[int][]models.CheckRecord, error) {
+	return nil, nil
+}
+func (m *mockStore) ExportData() (models.Backup, error) { return models.Backup{}, nil }
+func (m *mockStore) ImportData(models.Backup) error     { return nil }
+
+func TestMetricsHandler(t *testing.T) {
+	ms := &mockStore{
+		sites: []models.Site{
+			{ID: 1, Name: "Example", URL: "https://example.com", Type: "http", Interval: 30},
+			{ID: 2, Name: "DNS Check", Type: "dns", Interval: 60},
+		},
+	}
+	eng := monitor.NewEngine(ms)
+	ctx, cancel := context.WithCancel(context.Background())
+	eng.Start(ctx)
+	time.Sleep(100 * time.Millisecond)
+
+	rec := httptest.NewRecorder()
+	Handler(eng)(rec, httptest.NewRequest("GET", "/metrics", nil))
+	cancel()
+
+	if rec.Code != http.StatusOK {
+		t.Fatalf("expected 200, got %d", rec.Code)
+	}
+
+	body := rec.Body.String()
+
+	ct := rec.Header().Get("Content-Type")
+	if !strings.Contains(ct, "text/plain") {
+		t.Errorf("expected text/plain content type, got %q", ct)
+	}
+
+	expected := []string{
+		"# HELP upkeep_monitor_up",
+		"# TYPE upkeep_monitor_up gauge",
+		`upkeep_monitor_up{id="1",name="Example",type="http"}`,
+		`upkeep_monitor_up{id="2",name="DNS Check",type="dns"}`,
+		"# HELP upkeep_monitor_latency_seconds",
+		"# HELP upkeep_monitor_paused",
+		"# HELP upkeep_monitor_checks_total",
+	}
+	for _, s := range expected {
+		if !strings.Contains(body, s) {
+			t.Errorf("missing expected line: %s", s)
+		}
+	}
+}
+
+func TestEscapeLabelValue(t *testing.T) {
+	cases := []struct{ in, want string }{
+		{`simple`, `simple`},
+		{`has "quotes"`, `has \"quotes\"`},
+		{"has\nnewline", `has\nnewline`},
+		{`back\slash`, `back\\slash`},
+	}
+	for _, tc := range cases {
+		got := escapeLabelValue(tc.in)
+		if got != tc.want {
+			t.Errorf("escapeLabelValue(%q) = %q, want %q", tc.in, got, tc.want)
+		}
+	}
+}

internal/server/server.go

@@ -4,6 +4,7 @@ import (
 	"encoding/json"
 	"fmt"
 	"go-upkeep/internal/importer"
+	"go-upkeep/internal/metrics"
 	"go-upkeep/internal/models"
 	"go-upkeep/internal/monitor"
 	"go-upkeep/internal/store"
@@ -242,7 +243,10 @@ func Start(cfg ServerConfig, s store.Store, eng *monitor.Engine) {
 		w.Write([]byte(fmt.Sprintf("Imported %d monitors, %d alerts from Kuma v%s", len(backup.Sites), len(backup.Alerts), kb.Version)))
 	})
 
-	// 6. Status Page
+	// 6. Prometheus Metrics
+	mux.HandleFunc("/metrics", metrics.Handler(eng))
+
+	// 7. Status Page
 	if cfg.EnableStatus {
 		mux.HandleFunc("/status", func(w http.ResponseWriter, r *http.Request) { renderStatusPage(w, cfg.Title, eng) })
 		mux.HandleFunc("/status/json", func(w http.ResponseWriter, r *http.Request) {