[INS-246] Add Google Gemini API key detector (#4649)

* add google gemini api key detector

* change detector name to google cloud api key, mark as verified if 403 is returned

* add build tags for integration test

* changes in defaults.go

* Revert "changes in defaults.go"

This reverts commit 12e7b6f4aa29ba04bb357e32a9eff549b263cccc.

* Revert "change detector name to google cloud api key, mark as verified if 403 is returned"

This reverts commit e46bb29b40ae7510b686d12463fb6a9f3b6490c0.

* revert google cloud api changes, change keyword to gemini, add extra field active_google_key

* use aizasy as keyword instead of gemini

* close response body after draining

* remove \b from regex to support keys that end with -

* add \b to the beginning

pkg/detectors/googlegemini/googlegemini.go

@@ -0,0 +1,108 @@
+package googlegemini
+
+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()
+	// A trailing word boundary is not added because the key can end
+	// with a hyphen, and having a trailing \b will not match such a key
+	keyPat = regexp.MustCompile(`\b(AIzaSy[A-Za-z0-9_-]{33})`)
+)
+
+func (s Scanner) getClient() *http.Client {
+	if s.client != nil {
+		return s.client
+	}
+
+	return defaultClient
+}
+
+// Keywords are used for efficiently pre-filtering chunks.
+func (s Scanner) Keywords() []string { return []string{"aizasy"} }
+
+func (s Scanner) Type() detectorspb.DetectorType {
+	return detectorspb.DetectorType_GoogleGeminiAPIKey
+}
+
+func (s Scanner) Description() string {
+	return "Google Gemini API provides access to Google's latest generative AI models for building applications that understand and generate text, images, audio, and code with high performance and low latency."
+}
+
+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: s.Type(),
+			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
+}
+
+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() {
+		_, _ = io.Copy(io.Discard, res.Body)
+		_ = res.Body.Close()
+	}()
+
+	switch res.StatusCode {
+	case http.StatusOK:
+		// Key is valid and has access to gemini
+		return true, map[string]string{"active_google_key": "true"}, nil
+	case http.StatusForbidden:
+		// Key is valid but does not have access to gemini
+		return false, map[string]string{"active_google_key": "true"}, nil
+	case http.StatusBadRequest:
+		// Key is invalid (expired, revoked)
+		return false, nil, nil
+	default:
+		return false, nil, fmt.Errorf("unexpected status code: %d", res.StatusCode)
+	}
+}

pkg/detectors/googlegemini/googlegemini_integration_test.go

@@ -0,0 +1,148 @@
+//go:build detectors
+// +build detectors
+
+package googlegemini
+
+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 TestGoogleGemini_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)
+	}
+
+	keyGemini := testSecrets.MustGetField("GOOGLE_GEMINI_API_KEY")
+	keyNonGemini := testSecrets.MustGetField("GOOGLE_CLOUD_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 gemini key %s within", keyGemini)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_GoogleGeminiAPIKey,
+					Verified:     true,
+					ExtraData: map[string]string{
+						"active_google_key": "true",
+					},
+				},
+			},
+			wantErr: false,
+		},
+		{
+			name: "found, unverified but valid google cloud api key",
+			s:    Scanner{},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a google api key %s within", keyNonGemini)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_GoogleGeminiAPIKey,
+					Verified:     false,
+					ExtraData: map[string]string{
+						"active_google_key": "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_GoogleGeminiAPIKey,
+					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("GoogleGeminiAPIKey.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("GoogleGeminiAPIKey.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/googlegemini/googlegemini_test.go

@@ -0,0 +1,80 @@
+package googlegemini
+
+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 TestGoogleGemini_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("gemini api key = '%s'", validPattern),
+			want:  []string{validPattern},
+		},
+		{
+			name:  "invalid pattern",
+			input: fmt.Sprintf("gemini 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/googlegemini"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/googleoauth2"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/grafana"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors/grafanaserviceaccount"
@@ -1209,6 +1210,7 @@ func buildDetectorList() []detectors.Detector {
 		&godaddyv1.Scanner{},
 		&godaddyv2.Scanner{},
 		&goodday.Scanner{},
+		&googlegemini.Scanner{},
 		&googleoauth2.Scanner{},
 		&grafana.Scanner{},
 		&grafanaserviceaccount.Scanner{},

pkg/pb/detectorspb/detectors.pb.go

@@ -1147,6 +1147,7 @@ const (
 	DetectorType_Photoroom                               DetectorType = 1038
 	DetectorType_JWT                                     DetectorType = 1039
 	DetectorType_OpenAIAdmin                             DetectorType = 1040
+	DetectorType_GoogleGeminiAPIKey                      DetectorType = 1041
 )
 
 // Enum value maps for DetectorType.
@@ -2189,6 +2190,7 @@ var (
 		1038: "Photoroom",
 		1039: "JWT",
 		1040: "OpenAIAdmin",
+		1041: "GoogleGeminiAPIKey",
 	}
 	DetectorType_value = map[string]int32{
 		"Alibaba":                               0,
@@ -3228,6 +3230,7 @@ var (
 		"Photoroom":                         1038,
 		"JWT":                               1039,
 		"OpenAIAdmin":                       1040,
+		"GoogleGeminiAPIKey":                1041,
 	}
 )
 
@@ -3681,7 +3684,7 @@ var file_detectors_proto_rawDesc = []byte{
 	0x4c, 0x41, 0x49, 0x4e, 0x10, 0x01, 0x12, 0x0a, 0x0a, 0x06, 0x42, 0x41, 0x53, 0x45, 0x36, 0x34,
 	0x10, 0x02, 0x12, 0x09, 0x0a, 0x05, 0x55, 0x54, 0x46, 0x31, 0x36, 0x10, 0x03, 0x12, 0x13, 0x0a,
 	0x0f, 0x45, 0x53, 0x43, 0x41, 0x50, 0x45, 0x44, 0x5f, 0x55, 0x4e, 0x49, 0x43, 0x4f, 0x44, 0x45,
-	0x10, 0x04, 0x2a, 0xd7, 0x86, 0x01, 0x0a, 0x0c, 0x44, 0x65, 0x74, 0x65, 0x63, 0x74, 0x6f, 0x72,
+	0x10, 0x04, 0x2a, 0xf0, 0x86, 0x01, 0x0a, 0x0c, 0x44, 0x65, 0x74, 0x65, 0x63, 0x74, 0x6f, 0x72,
 	0x54, 0x79, 0x70, 0x65, 0x12, 0x0b, 0x0a, 0x07, 0x41, 0x6c, 0x69, 0x62, 0x61, 0x62, 0x61, 0x10,
 	0x00, 0x12, 0x08, 0x0a, 0x04, 0x41, 0x4d, 0x51, 0x50, 0x10, 0x01, 0x12, 0x07, 0x0a, 0x03, 0x41,
 	0x57, 0x53, 0x10, 0x02, 0x12, 0x09, 0x0a, 0x05, 0x41, 0x7a, 0x75, 0x72, 0x65, 0x10, 0x03, 0x12,
@@ -4758,12 +4761,13 @@ var file_detectors_proto_rawDesc = []byte{
 	0x61, 0x73, 0x65, 0x41, 0x63, 0x63, 0x65, 0x73, 0x73, 0x54, 0x6f, 0x6b, 0x65, 0x6e, 0x10, 0x8d,
 	0x08, 0x12, 0x0e, 0x0a, 0x09, 0x50, 0x68, 0x6f, 0x74, 0x6f, 0x72, 0x6f, 0x6f, 0x6d, 0x10, 0x8e,
 	0x08, 0x12, 0x08, 0x0a, 0x03, 0x4a, 0x57, 0x54, 0x10, 0x8f, 0x08, 0x12, 0x10, 0x0a, 0x0b, 0x4f,
-	0x70, 0x65, 0x6e, 0x41, 0x49, 0x41, 0x64, 0x6d, 0x69, 0x6e, 0x10, 0x90, 0x08, 0x42, 0x3d, 0x5a,
+	0x70, 0x65, 0x6e, 0x41, 0x49, 0x41, 0x64, 0x6d, 0x69, 0x6e, 0x10, 0x90, 0x08, 0x12, 0x17, 0x0a,
-	0x3b, 0x67, 0x69, 0x74, 0x68, 0x75, 0x62, 0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x74, 0x72, 0x75, 0x66,
+	0x12, 0x47, 0x6f, 0x6f, 0x67, 0x6c, 0x65, 0x47, 0x65, 0x6d, 0x69, 0x6e, 0x69, 0x41, 0x50, 0x49,
-	0x66, 0x6c, 0x65, 0x73, 0x65, 0x63, 0x75, 0x72, 0x69, 0x74, 0x79, 0x2f, 0x74, 0x72, 0x75, 0x66,
+	0x4b, 0x65, 0x79, 0x10, 0x91, 0x08, 0x42, 0x3d, 0x5a, 0x3b, 0x67, 0x69, 0x74, 0x68, 0x75, 0x62,
-	0x66, 0x6c, 0x65, 0x68, 0x6f, 0x67, 0x2f, 0x76, 0x33, 0x2f, 0x70, 0x6b, 0x67, 0x2f, 0x70, 0x62,
+	0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x73, 0x65, 0x63, 0x75,
-	0x2f, 0x64, 0x65, 0x74, 0x65, 0x63, 0x74, 0x6f, 0x72, 0x73, 0x70, 0x62, 0x62, 0x06, 0x70, 0x72,
+	0x72, 0x69, 0x74, 0x79, 0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x68, 0x6f, 0x67, 0x2f,
-	0x6f, 0x74, 0x6f, 0x33,
+	0x76, 0x33, 0x2f, 0x70, 0x6b, 0x67, 0x2f, 0x70, 0x62, 0x2f, 0x64, 0x65, 0x74, 0x65, 0x63, 0x74,
+	0x6f, 0x72, 0x73, 0x70, 0x62, 0x62, 0x06, 0x70, 0x72, 0x6f, 0x74, 0x6f, 0x33,
 }
 
 var (

proto/detectors.proto

@@ -1050,6 +1050,7 @@ enum DetectorType {
   Photoroom = 1038;
   JWT = 1039;
   OpenAIAdmin = 1040;
+  GoogleGeminiAPIKey = 1041;
 }
 
 message Result {