From b9ea2f5d4ee73c79b305290a1f2e249fafb0159e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=C4=81h=CC=95=CC=B3m=CD=AD=CD=AD=CD=A8=CD=A9=CC=90e=CD=AC?= =?UTF-8?q?=CC=81=CD=8B=CD=AC=CC=8A=CC=93=CD=82=CC=98d?= <13666360+0x1@users.noreply.github.com> Date: Wed, 22 May 2024 15:46:14 -0400 Subject: [PATCH] adding Groq detector (#2873) * adding Groq detector * using prefix as keyword --- pkg/detectors/groq/groq.go | 97 +++++++++++++ pkg/detectors/groq/groq_test.go | 221 +++++++++++++++++++++++++++++ pkg/engine/defaults.go | 2 + pkg/pb/detectorspb/detectors.pb.go | 17 ++- proto/detectors.proto | 1 + 5 files changed, 331 insertions(+), 7 deletions(-) create mode 100644 pkg/detectors/groq/groq.go create mode 100644 pkg/detectors/groq/groq_test.go diff --git a/pkg/detectors/groq/groq.go b/pkg/detectors/groq/groq.go new file mode 100644 index 000000000000..c5d52ea51d53 --- /dev/null +++ b/pkg/detectors/groq/groq.go @@ -0,0 +1,97 @@ +package groq + +import ( + "context" + "fmt" + "io" + "net/http" + + 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 interface at compile time. +var _ detectors.Detector = (*Scanner)(nil) + +var ( + defaultClient = common.SaneHttpClient() + keyPat = regexp.MustCompile(`\b(gsk_[a-zA-Z0-9]{52})\b`) +) + +func (s Scanner) Keywords() []string { + return []string{"gsk_"} +} + +// FromData will find and optionally verify Groq secrets in a given set of bytes. +func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (results []detectors.Result, err error) { + dataStr := string(data) + + uniqueMatches := make(map[string]struct{}) + for _, match := range keyPat.FindAllStringSubmatch(dataStr, -1) { + uniqueMatches[match[1]] = struct{}{} + } + + for match := range uniqueMatches { + s1 := detectors.Result{ + DetectorType: detectorspb.DetectorType_Groq, + Raw: []byte(match), + } + + if verify { + client := s.client + if client == nil { + client = defaultClient + } + + isVerified, extraData, verificationErr := verifyMatch(ctx, client, match) + s1.Verified = isVerified + s1.ExtraData = extraData + s1.SetVerificationError(verificationErr, match) + } + + results = append(results, s1) + } + + return +} + +func verifyMatch(ctx context.Context, client *http.Client, token string) (bool, map[string]string, error) { + // https://console.groq.com/docs/models + // This endpoint will return a JSON list of all active models. + req, err := http.NewRequestWithContext(ctx, http.MethodGet, "https://api.groq.com/openai/v1/models", nil) + if err != nil { + return false, nil, nil + } + req.Header.Add("Authorization", "Bearer "+token) + + res, err := client.Do(req) + if err != nil { + return false, nil, err + } + defer func() { + _, _ = io.Copy(io.Discard, res.Body) + _ = res.Body.Close() + }() + + switch res.StatusCode { + case http.StatusOK: + // If the endpoint returns useful information, we can return it as a map. + return true, nil, nil + case http.StatusUnauthorized: + // The secret is determinately not verified (nothing to do) + return false, nil, nil + default: + return false, nil, fmt.Errorf("unexpected HTTP response status %d", res.StatusCode) + } +} + +func (s Scanner) Type() detectorspb.DetectorType { + return detectorspb.DetectorType_Groq +} diff --git a/pkg/detectors/groq/groq_test.go b/pkg/detectors/groq/groq_test.go new file mode 100644 index 000000000000..5a836bb8055b --- /dev/null +++ b/pkg/detectors/groq/groq_test.go @@ -0,0 +1,221 @@ +//go:build detectors +// +build detectors + +package groq + +import ( + "context" + "fmt" + "testing" + "time" + + "github.com/google/go-cmp/cmp" + "github.com/google/go-cmp/cmp/cmpopts" + + "github.com/trufflesecurity/trufflehog/v3/pkg/common" + "github.com/trufflesecurity/trufflehog/v3/pkg/detectors" + "github.com/trufflesecurity/trufflehog/v3/pkg/engine/ahocorasick" + "github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb" +) + +func TestGroq_Pattern(t *testing.T) { + d := Scanner{} + ahoCorasickCore := ahocorasick.NewAhoCorasickCore([]detectors.Detector{d}) + tests := []struct { + name string + input string + want []string + }{ + { + name: "typical pattern", + input: "groq_token = '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'", + want: []string{"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"}, + }, + } + + for _, test := range tests { + t.Run(test.name, func(t *testing.T) { + chunkSpecificDetectors := make(map[ahocorasick.DetectorKey]detectors.Detector, 2) + ahoCorasickCore.PopulateMatchingDetectors(test.input, chunkSpecificDetectors) + if len(chunkSpecificDetectors) == 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) + } + }) + } +} + +func TestGroq_FromChunk(t *testing.T) { + ctx, cancel := context.WithTimeout(context.Background(), time.Second*5) + defer cancel() + testSecrets, err := common.GetSecret(ctx, "trufflehog-testing", "detectors5") + if err != nil { + t.Fatalf("could not get test secrets from GCP: %s", err) + } + secret := testSecrets.MustGetField("GROQ") + inactiveSecret := testSecrets.MustGetField("GROQ_INACTIVE") + + type args struct { + ctx context.Context + data []byte + verify bool + } + tests := []struct { + name string + s Scanner + args args + want []detectors.Result + wantErr bool + wantVerificationErr bool + }{ + { + name: "found, verified", + s: Scanner{}, + args: args{ + ctx: context.Background(), + data: []byte(fmt.Sprintf("You can find a groq secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detectorspb.DetectorType_Groq, + Verified: true, + }, + }, + wantErr: false, + wantVerificationErr: false, + }, + { + name: "found, unverified", + s: Scanner{}, + args: args{ + ctx: context.Background(), + data: []byte(fmt.Sprintf("You can find a groq secret %s within but not valid", inactiveSecret)), // the secret would satisfy the regex but not pass validation + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detectorspb.DetectorType_Groq, + Verified: false, + }, + }, + wantErr: false, + wantVerificationErr: 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, + wantVerificationErr: false, + }, + { + name: "found, would be verified if not for timeout", + s: Scanner{client: common.SaneHttpClientTimeOut(1 * time.Microsecond)}, + args: args{ + ctx: context.Background(), + data: []byte(fmt.Sprintf("You can find a groq secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detectorspb.DetectorType_Groq, + Verified: false, + }, + }, + wantErr: false, + wantVerificationErr: true, + }, + { + name: "found, verified but unexpected api surface", + s: Scanner{client: common.ConstantResponseHttpClient(404, "")}, + args: args{ + ctx: context.Background(), + data: []byte(fmt.Sprintf("You can find a groq secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detectorspb.DetectorType_Groq, + Verified: false, + }, + }, + wantErr: false, + wantVerificationErr: true, + }, + } + for _, tt := range tests { + t.Run(tt.name, func(t *testing.T) { + got, err := tt.s.FromData(tt.args.ctx, tt.args.verify, tt.args.data) + if (err != nil) != tt.wantErr { + t.Errorf("Groq.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]) + } + if (got[i].VerificationError() != nil) != tt.wantVerificationErr { + t.Fatalf("wantVerificationError = %v, verification error = %v", tt.wantVerificationErr, got[i].VerificationError()) + } + } + ignoreOpts := cmpopts.IgnoreFields(detectors.Result{}, "Raw", "verificationError") + if diff := cmp.Diff(got, tt.want, ignoreOpts); diff != "" { + t.Errorf("Groq.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) + } + } + }) + } +} diff --git a/pkg/engine/defaults.go b/pkg/engine/defaults.go index 612750ed5ac0..8deea1def9ab 100644 --- a/pkg/engine/defaults.go +++ b/pkg/engine/defaults.go @@ -311,6 +311,7 @@ import ( "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/graphcms" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/graphhopper" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/groovehq" + "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/groq" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/gtmetrix" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/guardianapi" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/gumroad" @@ -1604,6 +1605,7 @@ func DefaultDetectors() []detectors.Detector { wiz.Scanner{}, onfleet.Scanner{}, intra42.Scanner{}, + groq.Scanner{}, } } diff --git a/pkg/pb/detectorspb/detectors.pb.go b/pkg/pb/detectorspb/detectors.pb.go index cdb00c92324e..ea499fc6b6f1 100644 --- a/pkg/pb/detectorspb/detectors.pb.go +++ b/pkg/pb/detectorspb/detectors.pb.go @@ -1087,6 +1087,7 @@ const ( DetectorType_Pagarme DetectorType = 985 DetectorType_Onfleet DetectorType = 986 DetectorType_Intra42 DetectorType = 987 + DetectorType_Groq DetectorType = 988 ) // Enum value maps for DetectorType. @@ -2076,6 +2077,7 @@ var ( 985: "Pagarme", 986: "Onfleet", 987: "Intra42", + 988: "Groq", } DetectorType_value = map[string]int32{ "Alibaba": 0, @@ -3062,6 +3064,7 @@ var ( "Pagarme": 985, "Onfleet": 986, "Intra42": 987, + "Groq": 988, } ) @@ -3515,7 +3518,7 @@ 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, 0x89, 0x7e, 0x0a, 0x0c, 0x44, 0x65, 0x74, 0x65, 0x63, 0x74, 0x6f, 0x72, 0x54, + 0x10, 0x04, 0x2a, 0x94, 0x7e, 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, 0x0a, @@ -4523,12 +4526,12 @@ var file_detectors_proto_rawDesc = []byte{ 0x65, 0x6e, 0x74, 0x69, 0x61, 0x6c, 0x73, 0x10, 0xd7, 0x07, 0x12, 0x08, 0x0a, 0x03, 0x57, 0x69, 0x7a, 0x10, 0xd8, 0x07, 0x12, 0x0c, 0x0a, 0x07, 0x50, 0x61, 0x67, 0x61, 0x72, 0x6d, 0x65, 0x10, 0xd9, 0x07, 0x12, 0x0c, 0x0a, 0x07, 0x4f, 0x6e, 0x66, 0x6c, 0x65, 0x65, 0x74, 0x10, 0xda, 0x07, - 0x12, 0x0c, 0x0a, 0x07, 0x49, 0x6e, 0x74, 0x72, 0x61, 0x34, 0x32, 0x10, 0xdb, 0x07, 0x42, 0x3d, - 0x5a, 0x3b, 0x67, 0x69, 0x74, 0x68, 0x75, 0x62, 0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x74, 0x72, 0x75, - 0x66, 0x66, 0x6c, 0x65, 0x73, 0x65, 0x63, 0x75, 0x72, 0x69, 0x74, 0x79, 0x2f, 0x74, 0x72, 0x75, - 0x66, 0x66, 0x6c, 0x65, 0x68, 0x6f, 0x67, 0x2f, 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, + 0x12, 0x0c, 0x0a, 0x07, 0x49, 0x6e, 0x74, 0x72, 0x61, 0x34, 0x32, 0x10, 0xdb, 0x07, 0x12, 0x09, + 0x0a, 0x04, 0x47, 0x72, 0x6f, 0x71, 0x10, 0xdc, 0x07, 0x42, 0x3d, 0x5a, 0x3b, 0x67, 0x69, 0x74, + 0x68, 0x75, 0x62, 0x2e, 0x63, 0x6f, 0x6d, 0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x73, + 0x65, 0x63, 0x75, 0x72, 0x69, 0x74, 0x79, 0x2f, 0x74, 0x72, 0x75, 0x66, 0x66, 0x6c, 0x65, 0x68, + 0x6f, 0x67, 0x2f, 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 ( diff --git a/proto/detectors.proto b/proto/detectors.proto index 65a916add54f..906f67525e63 100644 --- a/proto/detectors.proto +++ b/proto/detectors.proto @@ -997,6 +997,7 @@ enum DetectorType { Pagarme = 985; Onfleet = 986; Intra42 = 987; + Groq = 988; } message Result {