diff --git a/pkg/detectors/googleoauth2/googleoauth2_access_token.go b/pkg/detectors/googleoauth2/googleoauth2_access_token.go new file mode 100644 index 000000000000..215e0c2fc10b --- /dev/null +++ b/pkg/detectors/googleoauth2/googleoauth2_access_token.go @@ -0,0 +1,140 @@ +package googleoauth2 + +import ( + "context" + "encoding/json" + "fmt" + "github.com/trufflesecurity/trufflehog/v3/pkg/common" + "github.com/trufflesecurity/trufflehog/v3/pkg/detectors" + "github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb" + "io" + "net/http" + "regexp" + "strconv" + "time" +) + +type Scanner struct { + client *http.Client +} + +// Ensure the Scanner satisfies the interface at compile time. +var _ detectors.Detector = (*Scanner)(nil) + +var ( + defaultClient = common.SaneHttpClient() + + // https://github.com/dart-lang/pub-dev/blob/21837fa606c7871f50725bf6172b9af0df1e616e/app/lib/account/default_auth_provider.dart#L462 + // https://github.com/GoogleChrome/developer.chrome.com/blob/51dd7dd5d510ed85d86f5a91cb8fde50b62351c7/site/en/docs/webstore/using_webstore_api/index.md?plain=1#L95 + keyPat = regexp.MustCompile(`\b(ya29\.(?i:[a-z0-9_-]{10,}))(?:[^a-z0-9_-]|\z)`) +) + +// 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{"ya29."} +} + +// FromData will find and optionally verify Googleoauth2 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) + + tokens := make(map[string]struct{}) + for _, matches := range keyPat.FindAllStringSubmatch(dataStr, -1) { + tokens[matches[1]] = struct{}{} + } + + for token := range tokens { + s1 := detectors.Result{ + DetectorType: s.Type(), + Raw: []byte(token), + } + + if verify { + if s.client == nil { + s.client = defaultClient + } + + verified, extraData, vErr := s.verify(ctx, token) + s1.Verified = verified + s1.ExtraData = extraData + s1.SetVerificationError(vErr) + } + + // This function will check false positives for common test words, but also it will make sure the key appears 'random' enough to be a real key. + if !s1.Verified && detectors.IsKnownFalsePositive(token, detectors.DefaultFalsePositives, true) { + continue + } + + results = append(results, s1) + } + return +} + +func (s Scanner) verify(ctx context.Context, token string) (bool, map[string]string, error) { + // Based on https://stackoverflow.com/a/66957524 + req, err := http.NewRequestWithContext(ctx, http.MethodGet, "https://www.googleapis.com/oauth2/v3/tokeninfo?access_token="+token, nil) + if err != nil { + return false, nil, err + } + + res, err := s.client.Do(req) + if err != nil { + return false, nil, err + } + defer func() { + _ = res.Body.Close() + _, _ = io.Copy(io.Discard, res.Body) + }() + + if res.StatusCode == http.StatusOK { + var token tokenInfo + if err := json.NewDecoder(res.Body).Decode(&token); err != nil { + return false, nil, fmt.Errorf("failed to decode response: %w", err) + } + + extraData := map[string]string{ + "Google_email": token.Email, + "Email_verified": token.EmailVerified, + "Scope": token.Scope, + "Access_type": token.AccessType, + } + + exp, err := strconv.ParseInt(token.Expiry, 10, 64) + if err == nil { + extraData["Expires_at"] = time.Unix(exp, 0).String() + } + return true, extraData, nil + } else if res.StatusCode == http.StatusBadRequest { + var errInfo errorInfo + if err := json.NewDecoder(res.Body).Decode(&errInfo); err != nil { + return false, nil, fmt.Errorf("failed to decode response: %w", err) + } + + if errInfo.Error == "Invalid Value" { + // Definitively false. + return false, nil, nil + } else { + return false, nil, fmt.Errorf("unexpected error description '%s' for %s", errInfo.Error, req.URL) + } + } else { + return false, nil, fmt.Errorf("unexpected respones %d for %s", res.StatusCode, req.URL) + } +} + +type tokenInfo struct { + Expiry string `json:"exp"` + Scope string `json:"scope"` + Email string `json:"email"` + EmailVerified string `json:"email_verified"` + AccessType string `json:"access_type"` +} + +// {"error_description": "Invalid Value"} +type errorInfo struct { + Error string `json:"error_description"` +} + +func (s Scanner) Type() detectorspb.DetectorType { + return detectorspb.DetectorType_GoogleOauth2 +} diff --git a/pkg/detectors/googleoauth2/googleoauth2_access_token_test.go b/pkg/detectors/googleoauth2/googleoauth2_access_token_test.go new file mode 100644 index 000000000000..c5a2c2559439 --- /dev/null +++ b/pkg/detectors/googleoauth2/googleoauth2_access_token_test.go @@ -0,0 +1,162 @@ +//go:build detectors +// +build detectors + +package googleoauth2 + +import ( + "context" + "fmt" + "testing" + "time" + + "github.com/google/go-cmp/cmp" + "github.com/google/go-cmp/cmp/cmpopts" + + "github.com/trufflesecurity/trufflehog/v3/pkg/detectors" + + "github.com/trufflesecurity/trufflehog/v3/pkg/common" + "github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb" +) + +func TestGoogleoauth2_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("GOOGLEOAUTH2") + inactiveSecret := testSecrets.MustGetField("GOOGLEOAUTH2_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 googleoauth2 secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detectorspb.DetectorType_GoogleOauth2, + Verified: true, + }, + }, + wantErr: false, + wantVerificationErr: false, + }, + { + name: "found, unverified", + s: Scanner{}, + args: args{ + ctx: context.Background(), + data: []byte(fmt.Sprintf("You can find a googleoauth2 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_GoogleOauth2, + 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 googleoauth2 secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detectorspb.DetectorType_GoogleOauth2, + 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 googleoauth2 secret %s within", secret)), + verify: true, + }, + want: []detectors.Result{ + { + DetectorType: detectorspb.DetectorType_GoogleOauth2, + 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("Googleoauth2.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("Googleoauth2.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 aec2ef61dce3..f459d05f1fe8 100644 --- a/pkg/engine/defaults.go +++ b/pkg/engine/defaults.go @@ -305,6 +305,7 @@ import ( "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/gocanvas" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/gocardless" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/goodday" + "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/googleoauth2" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/grafana" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/grafanaserviceaccount" "github.com/trufflesecurity/trufflehog/v3/pkg/detectors/graphcms" @@ -1600,6 +1601,7 @@ func DefaultDetectors() []detectors.Detector { azurestorage.Scanner{}, planetscaledb.Scanner{}, jiratoken_v2.Scanner{}, + &googleoauth2.Scanner{}, } }