[INS-246] Add Google Cloud API key detector

pkg/detectors/googlecloudapikey/googlecloudapikey.go

@@ -0,0 +1,108 @@
+package googlecloudapikey
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"net/http"
+	"strings"
+
+	regexp "github.com/wasilibs/go-re2"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
+)
+
+type Scanner struct {
+	client *http.Client
+}
+
+// Ensure the Scanner satisfies the interfaces at compile time.
+var _ detectors.Detector = (*Scanner)(nil)
+
+var (
+	defaultClient = common.SaneHttpClient()
+	keyPat        = regexp.MustCompile(`\b(AIza[A-Za-z0-9_-]{35})\b`)
+)
+
+func (s Scanner) getClient() *http.Client {
+	if s.client != nil {
+		return s.client
+	}
+
+	return defaultClient
+}
+
+// Keywords are used for efficiently pre-filtering chunks.
+// Use identifiers in the secret preferably, or the provider name.
+func (s Scanner) Keywords() []string { return []string{"aiza"} }
+
+func (s Scanner) Type() detectorspb.DetectorType {
+	return detectorspb.DetectorType_GoogleCloudAPIKey
+}
+
+func (s Scanner) Description() string {
+	return "Google Cloud API Key provides access to Google Cloud services."
+}
+
+func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (results []detectors.Result, err error) {
+	dataStr := string(data)
+
+	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	for _, match := range matches {
+		resMatch := strings.TrimSpace(match[1])
+
+		s1 := detectors.Result{
+			DetectorType: detectorspb.DetectorType_GoogleCloudAPIKey,
+			Raw:          []byte(resMatch),
+			Redacted:     resMatch[:8] + "...",
+		}
+
+		if verify {
+			isVerified, extraData, verificationErr := s.verify(ctx, resMatch)
+			s1.Verified = isVerified
+			s1.ExtraData = extraData
+			s1.SetVerificationError(verificationErr)
+		}
+
+		results = append(results, s1)
+	}
+
+	return results, nil
+}
+
+// verifies the provide google api key using the gemini /v1/models endpoint
+// 200 response indicates that key is live with gemini access
+// 403 indicates that key is live, but restricted or not enabled for gemini
+// 400 indicates that the key is inactive (invalid, expired or rotated)
+func (s Scanner) verify(ctx context.Context, key string) (bool, map[string]string, error) {
+	req, err := http.NewRequestWithContext(
+		ctx, http.MethodGet, "https://generativelanguage.googleapis.com/v1/models", http.NoBody)
+	if err != nil {
+		return false, nil, fmt.Errorf("error constructing request: %w", err)
+	}
+	req.Header.Add("Content-Type", "application/json")
+	req.Header.Add("X-goog-api-key", key)
+
+	client := s.getClient()
+	res, err := client.Do(req)
+	if err != nil {
+		return false, nil, fmt.Errorf("error making request: %w", err)
+	}
+	defer func() {
+		_ = res.Body.Close()
+		_, _ = io.Copy(io.Discard, res.Body)
+	}()
+
+	switch res.StatusCode {
+	case http.StatusOK:
+		return true, map[string]string{"gemini_enabled": "true"}, nil
+	case http.StatusForbidden:
+		return true, map[string]string{"gemini_enabled": "false"}, nil
+	case http.StatusBadRequest:
+		return false, nil, nil
+	default:
+		return false, nil, fmt.Errorf("unexpected status code: %d", res.StatusCode)
+	}
+}

pkg/detectors/googlecloudapikey/googlecloudapikey_integration_test.go

@@ -0,0 +1,148 @@
+//go:build detectors
+// +build detectors
+
+package googlecloudapikey
+
+import (
+	"context"
+	"fmt"
+	"testing"
+	"time"
+
+	"github.com/kylelemons/godebug/pretty"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
+)
+
+func TestGoogleCloudAPIKey_FromChunk(t *testing.T) {
+	ctx, cancel := context.WithTimeout(context.Background(), time.Second*5)
+	defer cancel()
+	testSecrets, err := common.GetSecret(ctx, "trufflehog-testing", "detectors6")
+	if err != nil {
+		t.Fatalf("could not get test secrets from GCP: %s", err)
+	}
+
+	key := testSecrets.MustGetField("GOOGLE_CLOUD_API_KEY")
+	keyGemini := testSecrets.MustGetField("GOOGLE_GEMINI_API_KEY")
+	keyInactive := testSecrets.MustGetField("GOOGLE_GEMINI_API_KEY_INACTIVE")
+
+	type args struct {
+		ctx    context.Context
+		data   []byte
+		verify bool
+	}
+	tests := []struct {
+		name    string
+		s       Scanner
+		args    args
+		want    []detectors.Result
+		wantErr bool
+	}{
+		{
+			name: "found, verified",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a google api key %s within", key)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_GoogleCloudAPIKey,
+					Verified:     true,
+					ExtraData: map[string]string{
+						"gemini_enabled": "false",
+					},
+				},
+			},
+			wantErr: false,
+		},
+		{
+			name: "found, verified with gemini enabled",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a google api key %s within", keyGemini)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_GoogleCloudAPIKey,
+					Verified:     true,
+					ExtraData: map[string]string{
+						"gemini_enabled": "true",
+					},
+				},
+			},
+			wantErr: false,
+		},
+		{
+			name: "found, unverified",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a google api key %s within", keyInactive)), // the secret would satisfy the regex but not pass validation
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_GoogleCloudAPIKey,
+					Verified:     false,
+				},
+			},
+			wantErr: false,
+		},
+		{
+			name: "not found",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte("You cannot find the secret within"),
+				verify: true,
+			},
+			want:    nil,
+			wantErr: false,
+		},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			s := Scanner{}
+			got, err := s.FromData(tt.args.ctx, tt.args.verify, tt.args.data)
+			if (err != nil) != tt.wantErr {
+				t.Errorf("GoogleCloudAPIKey.FromData() error = %v, wantErr %v", err, tt.wantErr)
+				return
+			}
+			for i := range got {
+				if len(got[i].Raw) == 0 {
+					t.Fatalf("no raw secret present: \n %+v", got[i])
+				}
+				got[i].Raw = nil
+				if len(got[i].Redacted) == 0 {
+					t.Fatalf("no redacted secret present: \n %+v", got[i])
+				}
+				got[i].Redacted = ""
+			}
+			if diff := pretty.Compare(got, tt.want); diff != "" {
+				t.Errorf("GoogleCloudAPIKey.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
+			}
+		})
+	}
+}
+
+func BenchmarkFromData(benchmark *testing.B) {
+	ctx := context.Background()
+	s := Scanner{}
+	for name, data := range detectors.MustGetBenchmarkData() {
+		benchmark.Run(name, func(b *testing.B) {
+			b.ResetTimer()
+			for n := 0; n < b.N; n++ {
+				_, err := s.FromData(ctx, false, data)
+				if err != nil {
+					b.Fatal(err)
+				}
+			}
+		})
+	}
+}

pkg/detectors/googlecloudapikey/googlecloudapikey_test.go

@@ -0,0 +1,80 @@
+package googlecloudapikey
+
+import (
+	"context"
+	"fmt"
+	"testing"
+
+	"github.com/google/go-cmp/cmp"
+
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/engine/ahocorasick"
+)
+
+var (
+	validPattern   = "AIzaSyDW1PXXav-TxriHUIrj1djZfHKIuInr7Aw"
+	invalidPattern = "AIzaSyDW1PXXav-TxriHUIrj1djZfHKIuInr7A"
+)
+
+func TestGoogleCloudAPIKey_Pattern(t *testing.T) {
+	d := Scanner{}
+	ahoCorasickCore := ahocorasick.NewAhoCorasickCore([]detectors.Detector{d})
+	tests := []struct {
+		name  string
+		input string
+		want  []string
+	}{
+		{
+			name:  "valid pattern",
+			input: fmt.Sprintf("google api key = '%s'", validPattern),
+			want:  []string{validPattern},
+		},
+		{
+			name:  "invalid pattern",
+			input: fmt.Sprintf("google api key = '%s'", invalidPattern),
+			want:  []string{},
+		},
+	}
+
+	for _, test := range tests {
+		t.Run(test.name, func(t *testing.T) {
+			matchedDetectors := ahoCorasickCore.FindDetectorMatches([]byte(test.input))
+			if len(matchedDetectors) == 0 {
+				t.Errorf("keywords '%v' not matched by: %s", d.Keywords(), test.input)
+				return
+			}
+
+			results, err := d.FromData(context.Background(), false, []byte(test.input))
+			if err != nil {
+				t.Errorf("error = %v", err)
+				return
+			}
+
+			if len(results) != len(test.want) {
+				if len(results) == 0 {
+					t.Errorf("did not receive result")
+				} else {
+					t.Errorf("expected %d results, only received %d", len(test.want), len(results))
+				}
+				return
+			}
+
+			actual := make(map[string]struct{}, len(results))
+			for _, r := range results {
+				if len(r.RawV2) > 0 {
+					actual[string(r.RawV2)] = struct{}{}
+				} else {
+					actual[string(r.Raw)] = struct{}{}
+				}
+			}
+			expected := make(map[string]struct{}, len(test.want))
+			for _, v := range test.want {
+				expected[v] = struct{}{}
+			}
+
+			if diff := cmp.Diff(expected, actual); diff != "" {
+				t.Errorf("%s diff: (-want +got)\n%s", test.name, diff)
+			}
+		})
+	}
+}

pkg/engine/defaults/defaults.go

@@ -337,6 +337,7 @@ import (
 	godaddyv1 "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/godaddy/v1"
 	godaddyv2 "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/godaddy/v2"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/goodday"
+	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/googlecloudapikey"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/googleoauth2"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/grafana"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/grafanaserviceaccount"
@@ -1208,6 +1209,7 @@ func buildDetectorList() []detectors.Detector {
 		&godaddyv1.Scanner{},
 		&godaddyv2.Scanner{},
 		&goodday.Scanner{},
+		&googlecloudapikey.Scanner{},
 		&googleoauth2.Scanner{},
 		&grafana.Scanner{},
 		&grafanaserviceaccount.Scanner{},