Merge branch 'feature-bytification' into detectors-9

* fix sentry detector to check response.

* use err.

* address comments.

pkg/detectors/dailyco/dailyco.go

@@ -1,11 +1,11 @@
 package dailyco
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,37 +14,30 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"daily"}) + `\b([0-9a-f]{64})\b`)
 )
 
-// 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{"daily"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("daily")}
 }
 
-// FromData will find and optionally verify DailyCO 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_DailyCO,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -52,14 +45,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", string(resMatch)))
 			res, err := client.Do(req)
-			if err == nil {
+			if err == nil && res != nil {
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/dandelion/dandelion.go

@@ -1,11 +1,11 @@
 package dandelion
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,51 +14,44 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
-	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"dandelion"}) + `\b([a-z0-9]{32})\b`)
+	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"dandelion"}) + `\b([0-9a-z]{32})\b`)
 )
 
-// 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{"dandelion"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("dandelion")}
 }
 
-// FromData will find and optionally verify Dandelion 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Dandelion,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://api.dandelion.eu/datatxt/li/v1/?text=Smart&token=%s", resMatch), nil)
+			req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://api.dandelion.eu/datatxt/li/v1/?text=Smart&token=%s", string(resMatch)), nil)
 			if err != nil {
 				continue
 			}
+			req.Header.Add("Content-Type", "application/json")
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/dareboost/dareboost.go

@@ -1,11 +1,11 @@
 package dareboost
 
 import (
+	"bytes"
 	"context"
 	"io"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,41 +14,35 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"dareboost"}) + `\b([0-9a-zA-Z]{60})\b`)
 )
 
-// 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{"dareboost"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("dareboost")}
 }
 
-// FromData will find and optionally verify Dareboost 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)
 
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Dareboost,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			payload := strings.NewReader(`{    "token": "` + resMatch + `",    "location": "Paris"}`)
+			payload := bytes.NewBuffer([]byte(`{    "token": "` + string(resMatch) + `",    "location": "Paris"}`))
 
 			req, err := http.NewRequestWithContext(ctx, "POST", "https://api.dareboost.com/0.8/config", payload)
 			if err != nil {
@@ -62,8 +56,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 					continue
 				}
 
-				bodyString := string(bodyBytes)
-				validResponse := strings.Contains(bodyString, `"status":200`)
+				validResponse := bytes.Contains(bodyBytes, []byte(`"status":200`))
 
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
@@ -73,7 +66,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 						s1.Verified = false
 					}
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/databox/databox.go

@@ -1,12 +1,12 @@
 package databox
 
 import (
+	"bytes"
 	"context"
 	b64 "encoding/base64"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -15,49 +15,42 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
-
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"databox"}) + common.BuildRegex(common.RegexPattern, "", 21))
 )
 
-// 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{"databox"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("databox")}
 }
 
-// FromData will find and optionally verify Databox 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)
 
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Databox,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 		if verify {
-			data := fmt.Sprintf("%s:", resMatch)
+			data := fmt.Sprintf("%s:", string(resMatch))
 			sEnc := b64.StdEncoding.EncodeToString([]byte(data))
-			payload := strings.NewReader(`{
+			payload := bytes.NewBuffer([]byte(`{
 				"data":[
 				  {
 					"$sales": 420,
 					"$visitors": 123000
 				  }
 				]
-			  }`)
+			  }`))
 
 			req, err := http.NewRequestWithContext(ctx, "POST", "https://push.databox.com", payload)
 			if err != nil {
@@ -72,7 +65,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/databrickstoken/databrickstoken.go

@@ -1,11 +1,11 @@
 package databrickstoken
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -27,48 +27,46 @@ var (
 
 // 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{"databricks", "dapi"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("databricks"), []byte("dapi")}
 }
 
 // FromData will find and optionally verify Databrickstoken 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
-	domainMatches := domain.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
+	domainMatches := domain.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		for _, domainmatch := range domainMatches {
 			if len(domainmatch) != 2 {
 				continue
 			}
-			resDomainMatch := strings.TrimSpace(domainmatch[1])
+
+			resDomainMatch := bytes.TrimSpace(domainmatch[1])
 
 			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_DatabricksToken,
-				Raw:          []byte(resMatch),
-				RawV2:        []byte(resMatch + resDomainMatch),
+				Raw:          resMatch,
+				RawV2:        append(resMatch, resDomainMatch...),
 			}
 
 			if verify {
-				req, err := http.NewRequestWithContext(ctx, "GET", resDomainMatch + "/api/2.0/clusters/list", nil)
+				req, err := http.NewRequestWithContext(ctx, "GET", string(resDomainMatch)+"/api/2.0/clusters/list", nil)
 				if err != nil {
 					continue
 				}
-				req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", resMatch))
+				req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", string(resMatch)))
 				res, err := client.Do(req)
 				if err == nil {
 					defer res.Body.Close()
 					if res.StatusCode >= 200 && res.StatusCode < 300 {
 						s1.Verified = true
 					} else {
-						// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 							continue
 						}

pkg/detectors/datadogtoken/datadogtoken.go

@@ -1,10 +1,10 @@
 package datadogtoken
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -13,108 +13,61 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
-	appPat = regexp.MustCompile(detectors.PrefixRegex([]string{"datadog", "dd"}) + `\b([a-zA-Z-0-9]{40})\b`)
+	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"datadog", "dd"}) + `\b([a-zA-Z-0-9]{40})\b`)
 	apiPat = regexp.MustCompile(detectors.PrefixRegex([]string{"datadog", "dd"}) + `\b([a-zA-Z-0-9]{32})\b`)
 )
 
-// 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{"datadog"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("datadog")}
 }
 
-// FromData will find and optionally verify DatadogToken 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)
+	matches := keyPat.FindAllSubmatch(data, -1)
+	apiMatches := apiPat.FindAllSubmatch(data, -1)
 
-	appMatches := appPat.FindAllStringSubmatch(dataStr, -1)
-	apiMatches := apiPat.FindAllStringSubmatch(dataStr, -1)
-
-	for _, apiMatch := range apiMatches {
-		if len(apiMatch) != 2 {
+	for _, match := range matches {
+		if len(match) != 2 {
 			continue
 		}
-		resApiMatch := strings.TrimSpace(apiMatch[1])
-		appIncluded := false
-		for _, appMatch := range appMatches {
-			if len(appMatch) != 2 {
+		resMatch := bytes.TrimSpace(match[1])
+		for _, apimatch := range apiMatches {
+			if len(apimatch) != 2 {
 				continue
 			}
-			resAppMatch := strings.TrimSpace(appMatch[1])
+			resApiMatch := bytes.TrimSpace(apimatch[1])
 
 			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_DatadogToken,
-				Raw:          []byte(resAppMatch),
-				RawV2:        []byte(resAppMatch + resApiMatch),
-				ExtraData: map[string]string{
-					"Type": "Application+APIKey",
-				},
+				Raw:          resMatch,
+				RawV2:        append(resMatch, resApiMatch...),
 			}
 
 			if verify {
-
 				req, err := http.NewRequestWithContext(ctx, "GET", "https://api.datadoghq.com/api/v2/users", nil)
 				if err != nil {
 					continue
 				}
 				req.Header.Add("Content-Type", "application/json")
-				req.Header.Add("DD-API-KEY", resApiMatch)
-				req.Header.Add("DD-APPLICATION-KEY", resAppMatch)
+				req.Header.Add("DD-API-KEY", string(resApiMatch))
+				req.Header.Add("DD-APPLICATION-KEY", string(resMatch))
 				res, err := client.Do(req)
 				if err == nil {
 					defer res.Body.Close()
 					if res.StatusCode >= 200 && res.StatusCode < 300 {
 						s1.Verified = true
 					} else {
-						// 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 detectors.IsKnownFalsePositive(resApiMatch, detectors.DefaultFalsePositives, true) {
+						if detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 							continue
 						}
 					}
 				}
 			}
-			appIncluded = true
-			results = append(results, s1)
-		}
 
-		if !appIncluded {
-			s1 := detectors.Result{
-				DetectorType: detectorspb.DetectorType_DatadogToken,
-				Raw:          []byte(resApiMatch),
-				RawV2:        []byte(resApiMatch),
-				ExtraData: map[string]string{
-					"Type": "APIKeyOnly",
-				},
-			}
-
-			if verify {
-
-				req, err := http.NewRequestWithContext(ctx, "GET", "https://api.datadoghq.com/api/v1/validate", nil)
-				if err != nil {
-					continue
-				}
-				req.Header.Add("Content-Type", "application/json")
-				req.Header.Add("DD-API-KEY", resApiMatch)
-				res, err := client.Do(req)
-				if err == nil {
-					defer res.Body.Close()
-					if res.StatusCode >= 200 && res.StatusCode < 300 {
-						s1.Verified = true
-					} else {
-						// 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 detectors.IsKnownFalsePositive(resApiMatch, detectors.DefaultFalsePositives, true) {
-							continue
-						}
-					}
-				}
-			}
 			results = append(results, s1)
 		}
 	}

pkg/detectors/datafire/datafire.go

@@ -1,11 +1,11 @@
 package datafire
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,37 +14,29 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
-
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"datafire"}) + `\b([a-z0-9\S]{175,190})\b`)
 )
 
-// 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{"datafire"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("datafire")}
 }
 
-// FromData will find and optionally verify DataFire 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_DataFire,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -52,14 +44,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", string(resMatch)))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/datagov/datagov.go

@@ -1,11 +1,11 @@
 package datagov
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,41 +14,33 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
-
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"data.gov"}) + `\b([a-zA-Z0-9]{40})\b`)
 )
 
-// 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{"data.gov"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("data.gov")}
 }
 
-// FromData will find and optionally verify DataGov 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_DataGov,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://api.ers.usda.gov/data/arms/state?api_key=%s", resMatch), nil)
+			req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://api.ers.usda.gov/data/arms/state?api_key=%s", string(resMatch)), nil)
 			if err != nil {
 				continue
 			}
@@ -58,7 +50,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}
@@ -68,7 +59,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 
 		results = append(results, s1)
 	}
-
 	return results, nil
 }
 

pkg/detectors/debounce/debounce.go

@@ -1,10 +1,10 @@
 package debounce
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -13,41 +13,37 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"debounce"}) + `\b([a-zA-Z0-9]{13})\b`)
 )
 
-// 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{"debounce"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("debounce")}
 }
 
-// FromData will find and optionally verify Debounce 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Debounce,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			req, err := http.NewRequestWithContext(ctx, "GET", "https://api.debounce.io/v1/?api="+resMatch+"&email=some@gmail.com", nil)
+			api := append([]byte("https://api.debounce.io/v1/?api="), resMatch...)
+			url := append(api, []byte("&email=some@gmail.com")...)
+
+			req, err := http.NewRequestWithContext(ctx, "GET", string(url), nil)
 			if err != nil {
 				continue
 			}
@@ -58,7 +54,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/deepai/deepai.go

@@ -7,7 +7,6 @@ import (
 	"mime/multipart"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -26,27 +25,22 @@ var (
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"deepai"}) + `\b([a-z0-9-]{36})\b`)
 )
 
-// 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{"deepai"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("deepai")}
 }
 
-// FromData will find and optionally verify DeepAI 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_DeepAI,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -56,7 +50,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			_, err = io.Copy(fw, strings.NewReader("test"))
+			_, err = io.Copy(fw, bytes.NewReader([]byte("test")))
 			if err != nil {
 				continue
 			}
@@ -66,15 +60,14 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				continue
 			}
 			req.Header.Add("Content-Type", writer.FormDataContentType())
-			req.Header.Add("api-key", resMatch)
+			req.Header.Add("api-key", string(resMatch))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+					if detectors.IsKnownFalsePositive([]byte(resMatch), detectors.DefaultFalsePositives, true) {
 						continue
 					}
 				}

pkg/detectors/deepgram/deepgram.go

@@ -1,11 +1,11 @@
 package deepgram
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,37 +14,29 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
-
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"deepgram"}) + `\b([0-9a-z]{40})\b`)
 )
 
-// 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{"deepgram"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("deepgram")}
 }
 
-// FromData will find and optionally verify Deepgram 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Deepgram,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -53,14 +45,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				continue
 			}
 			req.Header.Add("Content-Type", "application/json")
-			req.Header.Add("Authorization", fmt.Sprintf("Token %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Token %s", string(resMatch)))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/delighted/delighted.go

@@ -1,12 +1,12 @@
 package delighted
 
 import (
+	"bytes"
 	"context"
 	b64 "encoding/base64"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -15,69 +15,56 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
-
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
-	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"delighted"}) + `\b([a-z0-9A-Z]{32})\b`)
+	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"delighted"}) + `\b([a-zA-Z0-9]{32})\b`)
 )
 
-// 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{"delighted"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("delighted")}
 }
 
-// FromData will find and optionally verify Delighted 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Delighted,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			data := fmt.Sprintf("%s:", resMatch)
-			sEnc := b64.StdEncoding.EncodeToString([]byte(data))
-			payload := strings.NewReader(`{
+			sEnc := b64.StdEncoding.EncodeToString(resMatch)
+			payload := bytes.NewReader([]byte(`{
 				"email": "jony@appleseed.com",
 				"properties": { "Purchase Experience": "Mobile App", "State": "CA" }
-				}`)
+				}`))
 			req, err := http.NewRequestWithContext(ctx, "POST", "https://api.delighted.com/v1/people.json", payload)
-			if err != nil {
-				continue
-			}
-			req.Header.Add("Content-Type", "application/json")
-			req.Header.Add("Authorization", fmt.Sprintf("Basic %s", sEnc))
-			res, err := client.Do(req)
 			if err == nil {
-				defer res.Body.Close()
-				if res.StatusCode >= 200 && res.StatusCode < 300 {
-					s1.Verified = true
-				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
-						continue
+				req.Header.Add("Content-Type", "application/json")
+				req.Header.Add("Authorization", fmt.Sprintf("Basic %s", sEnc))
+				res, err := client.Do(req)
+				if err == nil {
+					defer res.Body.Close()
+					if res.StatusCode >= 200 && res.StatusCode < 300 {
+						s1.Verified = true
+					} else {
+						if detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+							continue
+						}
 					}
 				}
 			}
 		}
-
 		results = append(results, s1)
 	}
-
 	return results, nil
 }
 

pkg/detectors/demio/demio.go

@@ -1,11 +1,11 @@
 package demio
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,70 +14,59 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives
 	keyPat    = regexp.MustCompile(detectors.PrefixRegex([]string{"demio"}) + `\b([a-z0-9A-Z]{32})\b`)
 	secretPat = regexp.MustCompile(detectors.PrefixRegex([]string{"demio"}) + `\b([a-z0-9A-Z]{10,20})\b`)
 )
 
-// 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{"demio"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("demio")}
 }
 
-// FromData will find and optionally verify Demio 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)
+	keyMatches := keyPat.FindAllSubmatch(data, -1)
+	secretMatches := secretPat.FindAllSubmatch(data, -1)
 
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
-	idMatches := secretPat.FindAllStringSubmatch(dataStr, -1)
-	for _, match := range matches {
+	for _, match := range keyMatches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
-		for _, idmatch := range idMatches {
-			if len(match) != 2 {
+		resMatch := bytes.TrimSpace(match[1])
+
+		for _, idmatch := range secretMatches {
+			if len(idmatch) != 2 {
 				continue
 			}
-			resIdMatch := strings.TrimSpace(idmatch[1])
+			resIdMatch := bytes.TrimSpace(idmatch[1])
 
 			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_Demio,
-				Raw:          []byte(resMatch),
+				Raw:          resMatch,
 			}
 
 			if verify {
-				url := fmt.Sprintf("https://my.demio.com/api/v1/ping/query?api_key=%s&api_secret=%s", resMatch, resIdMatch)
+				url := fmt.Sprintf("https://my.demio.com/api/v1/ping/query?api_key=%s&api_secret=%s", string(resMatch), string(resIdMatch))
+
 				req, err := http.NewRequestWithContext(ctx, "GET", url, nil)
 				if err != nil {
 					continue
 				}
 				res, err := client.Do(req)
-				if err == nil {
-					defer res.Body.Close()
-					if res.StatusCode >= 200 && res.StatusCode < 300 {
-						s1.Verified = true
-					} else {
-						// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
-							continue
-						}
+				if err != nil || res.StatusCode < 200 || res.StatusCode >= 300 {
+					if detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+						continue
 					}
+				} else {
+					s1.Verified = true
 				}
 			}
-
 			results = append(results, s1)
 		}
-
 	}
-
 	return results, nil
 }
 

pkg/detectors/deputy/deputy.go

@@ -1,11 +1,11 @@
 package deputy
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,64 +14,58 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"deputy"}) + `\b([0-9a-z]{32})\b`)
 	urlPat = regexp.MustCompile(`\b([0-9a-z]{1,}\.as\.deputy\.com)\b`)
 )
 
-// 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{"deputy"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("deputy")}
 }
 
-// FromData will find and optionally verify Deputy 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)
 
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
-	urlMatches := urlPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
+	urlMatches := urlPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		for _, urlMatch := range urlMatches {
 			if len(urlMatch) != 2 {
 				continue
 			}
-			resURL := strings.TrimSpace(urlMatch[1])
+			resURL := bytes.TrimSpace(urlMatch[1])
 
 			s1 := detectors.Result{
 				DetectorType: detectorspb.DetectorType_Deputy,
-				Raw:          []byte(resMatch),
+				Raw:          resMatch,
 			}
 
 			if verify {
-				req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://%s/api/v1/me", resURL), nil)
+				req, err := http.NewRequestWithContext(ctx, "GET", fmt.Sprintf("https://%s/api/v1/me", string(resURL)), nil)
 				if err != nil {
 					continue
 				}
 				req.Header.Add("Content-Type", "application/json")
-				req.Header.Add("Authorization", fmt.Sprintf("OAuth %s", resMatch))
+				req.Header.Add("Authorization", fmt.Sprintf("OAuth %s", string(resMatch)))
 				res, err := client.Do(req)
-				if err == nil {
-					defer res.Body.Close()
-					if res.StatusCode >= 200 && res.StatusCode < 300 {
-						s1.Verified = true
-					} else {
-						// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
-							continue
-						}
+				if err != nil {
+					continue
+				}
+				defer res.Body.Close()
+				if res.StatusCode >= 200 && res.StatusCode < 300 {
+					s1.Verified = true
+				} else {
+					if detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+						continue
 					}
 				}
 			}
@@ -79,6 +73,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			results = append(results, s1)
 		}
 	}
+
 	return results, nil
 }
 

pkg/detectors/detectify/detectify.go

@@ -1,10 +1,10 @@
 package detectify
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -13,37 +13,30 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"detectify"}) + `\b([0-9a-z]{32})\b`)
 )
 
-// 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{"detectify"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("detectify")}
 }
 
-// FromData will find and optionally verify Detectify 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Detectify,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -51,14 +44,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("X-Detectify-Key", resMatch)
+			req.Header.Add("X-Detectify-Key", string(resMatch))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/detectlanguage/detectlanguage.go

@@ -1,11 +1,11 @@
 package detectlanguage
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,37 +14,30 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"detectlanguage"}) + `\b([a-z0-9]{32})\b`)
 )
 
-// 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{"detectlanguage"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("detectlanguage")}
 }
 
-// FromData will find and optionally verify DetectLanguage 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_DetectLanguage,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -52,14 +45,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err != nil {
 				continue
 			}
-			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", string(resMatch)))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/detectors.go

@@ -20,7 +20,7 @@ type Detector interface {
 	FromData(ctx context.Context, verify bool, data []byte) ([]Result, error)
 	// Keywords are used for efficiently pre-filtering chunks using substring operations.
 	// Use unique identifiers that are part of the secret if you can, or the provider name.
-	Keywords() []string
+	Keywords() [][]byte
 	// Type returns the DetectorType number from detectors.proto for the given detector.
 	Type() detectorspb.DetectorType
 }

pkg/detectors/dfuse/dfuse.go

@@ -1,10 +1,10 @@
 package dfuse
 
 import (
+	"bytes"
 	"context"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -25,29 +25,27 @@ var (
 
 // 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{"dfuse"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("dfuse")}
 }
 
 // FromData will find and optionally verify Dfuse 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Dfuse,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
-			payload := strings.NewReader(`{"api_key":"` + resMatch + `"}`)
+			payload := bytes.NewBuffer([]byte(`{"api_key":"` + string(resMatch) + `"}`))
 			req, err := http.NewRequestWithContext(ctx, "POST", "https://auth.dfuse.io/v1/auth/issue", payload)
 			if err != nil {
 				continue
@@ -59,7 +57,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/diffbot/diffbot.go

@@ -1,12 +1,12 @@
 package diffbot
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"io"
 	"net/http"
 	"regexp"
-	"strings"
 	"time"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
@@ -16,43 +16,36 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
 
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"diffbot"}) + `\b([a-z0-9]{32})\b`)
 )
 
-// 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{"diffbot"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("diffbot")}
 }
 
-// FromData will find and optionally verify Diffbot 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)
-
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Diffbot,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
 			timeout := 10 * time.Second
 			client.Timeout = timeout
-			req, err := http.NewRequest("GET", fmt.Sprintf("https://api.diffbot.com/v4/account?token=%s", resMatch), nil)
+			req, err := http.NewRequest("GET", fmt.Sprintf("https://api.diffbot.com/v4/account?token=%s", string(resMatch)), nil)
 			if err != nil {
 				continue
 			}
@@ -60,8 +53,7 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 			if err == nil {
 				bodyBytes, err := io.ReadAll(res.Body)
 				if err == nil {
-					bodyString := string(bodyBytes)
-					validResponse := strings.Contains(bodyString, `"token":`) && strings.Contains(bodyString, `"planCredits":`)
+					validResponse := bytes.Contains(bodyBytes, []byte(`"token":`)) && bytes.Contains(bodyBytes, []byte(`"planCredits":`))
 					defer res.Body.Close()
 					if res.StatusCode >= 200 && res.StatusCode < 300 {
 						if validResponse {
@@ -70,7 +62,6 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 							s1.Verified = false
 						}
 					} else {
-						// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 							continue
 						}

pkg/detectors/diggernaut/diggernaut.go

@@ -1,11 +1,11 @@
 package diggernaut
 
 import (
+	"bytes"
 	"context"
 	"fmt"
 	"net/http"
 	"regexp"
-	"strings"
 
 	"github.com/trufflesecurity/trufflehog/v3/pkg/common"
 	"github.com/trufflesecurity/trufflehog/v3/pkg/detectors"
@@ -14,37 +14,30 @@ import (
 
 type Scanner struct{}
 
-// Ensure the Scanner satisfies the interface at compile time
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
 	client = common.SaneHttpClient()
-
-	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"diggernaut"}) + `\b([0-9a-z]{40})\b`)
 )
 
-// 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{"diggernaut"}
+func (s Scanner) Keywords() [][]byte {
+	return [][]byte{[]byte("diggernaut")}
 }
 
-// FromData will find and optionally verify Diggernaut 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)
 
-	matches := keyPat.FindAllStringSubmatch(dataStr, -1)
+	matches := keyPat.FindAllSubmatch(data, -1)
 
 	for _, match := range matches {
 		if len(match) != 2 {
 			continue
 		}
-		resMatch := strings.TrimSpace(match[1])
+		resMatch := bytes.TrimSpace(match[1])
 
 		s1 := detectors.Result{
 			DetectorType: detectorspb.DetectorType_Diggernaut,
-			Raw:          []byte(resMatch),
+			Raw:          resMatch,
 		}
 
 		if verify {
@@ -53,14 +46,13 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 				continue
 			}
 			req.Header.Add("Accept", "application/json")
-			req.Header.Add("Authorization", fmt.Sprintf("Token %s", resMatch))
+			req.Header.Add("Authorization", fmt.Sprintf("Token %s", string(resMatch)))
 			res, err := client.Do(req)
 			if err == nil {
 				defer res.Body.Close()
 				if res.StatusCode >= 200 && res.StatusCode < 300 {
 					s1.Verified = true
 				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
 						continue
 					}

pkg/detectors/sentrytoken/sentrytoken.go

@@ -2,7 +2,10 @@ package sentrytoken
 
 import (
 	"context"
+	"encoding/json"
+	"errors"
 	"fmt"
+	"io"
 	"net/http"
 	"regexp"
 	"strings"
@@ -12,16 +15,20 @@ import (
 	"github.com/trufflesecurity/trufflehog/v3/pkg/pb/detectorspb"
 )
 
-type Scanner struct{}
+type Scanner struct {
+	client *http.Client
+}
 
 // Ensure the Scanner satisfies the interface at compile time.
 var _ detectors.Detector = (*Scanner)(nil)
 
 var (
-	client = common.SaneHttpClient()
+	defaultClient = common.SaneHttpClient()
 
 	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
 	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"sentry"}) + `\b([a-f0-9]{64})\b`)
+
+	errUnauthorized = fmt.Errorf("token unauthorized")
 )
 
 // Keywords are used for efficiently pre-filtering chunks.
@@ -47,32 +54,78 @@ func (s Scanner) FromData(ctx context.Context, verify bool, data []byte) (result
 		}
 
 		if verify {
-			req, err := http.NewRequestWithContext(ctx, "GET", "https://sentry.io/api/0/projects/", nil)
-			if err != nil {
-				continue
+			client := s.client
+			if client == nil {
+				client = defaultClient
 			}
-			req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", resMatch))
-			// req.Header.Add("Cookie", "sc=g2ceDPcsBwPLGXugHUv0coGBi09TQYKqmtp0u4JYdIvOTFd28yVvn1yiuXgmJJa8; session=.eJwNyE0OwiAQQGGsPwtP4kqoGQa4gUdgZ2CGiaZJCUqjGxOPbnfve7_h29QpHpVSvbw61To9StvEwzre9TkVbkPcr7jN5dPbNl7X1jUt_a4vLpEHgUycQSgFZ0YBRkGxnthxwIyBAgdngcF4YxBssiMjekGv2245_wHyySMG:1m0J5S:FjpRlBVuFx577UQp6GcSGWTq0uk")
-			res, err := client.Do(req)
-			if err == nil {
-				defer res.Body.Close()
-				if res.StatusCode >= 200 && res.StatusCode < 300 {
-					s1.Verified = true
-				} else {
-					// 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 detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
-						continue
-					}
-				}
+			isVerified, verificationErr := verifyToken(ctx, client, resMatch)
+
+			switch {
+			case errors.Is(verificationErr, errUnauthorized):
+				s1.Verified = false
+			case isVerified:
+				s1.Verified = true
+			default:
+				s1.VerificationError = verificationErr
 			}
 		}
 
+		if !s1.Verified && detectors.IsKnownFalsePositive(resMatch, detectors.DefaultFalsePositives, true) {
+			continue
+		}
 		results = append(results, s1)
 	}
 
 	return results, nil
 }
 
+type Response []Project
+
+type Project struct {
+	Organization Organization `json:"organization"`
+}
+
+type Organization struct {
+	ID   string `json:"id"`
+	Name string `json:"name"`
+}
+
+func verifyToken(ctx context.Context, client *http.Client, token string) (bool, error) {
+	req, err := http.NewRequestWithContext(ctx, http.MethodGet, "https://sentry.io/api/0/projects/", nil)
+	if err != nil {
+		return false, err
+	}
+	req.Header.Add("Authorization", fmt.Sprintf("Bearer %s", token))
+
+	res, err := client.Do(req)
+	if err != nil {
+		return false, err
+	}
+	defer res.Body.Close()
+
+	var isVerified bool
+	switch res.StatusCode {
+	case http.StatusOK, http.StatusForbidden:
+		isVerified = true
+	case http.StatusUnauthorized:
+		return false, errUnauthorized
+	default:
+		return false, fmt.Errorf("unexpected HTTP response status %d", res.StatusCode)
+	}
+
+	bytes, readErr := io.ReadAll(res.Body)
+	if readErr != nil {
+		return false, readErr
+	}
+
+	var resp Response
+	if err = json.Unmarshal(bytes, &resp); err != nil {
+		return false, err
+	}
+
+	return isVerified, err
+}
+
 func (s Scanner) Type() detectorspb.DetectorType {
 	return detectorspb.DetectorType_SentryToken
 }

pkg/detectors/sentrytoken/sentrytoken_test.go

@@ -9,7 +9,9 @@ import (
 	"testing"
 	"time"
 
-	"github.com/kylelemons/godebug/pretty"
+	"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"
@@ -32,11 +34,12 @@ func TestSentryToken_FromChunk(t *testing.T) {
 		verify bool
 	}
 	tests := []struct {
-		name    string
-		s       Scanner
-		args    args
-		want    []detectors.Result
-		wantErr bool
+		name                string
+		s                   Scanner
+		args                args
+		want                []detectors.Result
+		wantErr             bool
+		wantVerificationErr bool
 	}{
 		{
 			name: "found, verified",
@@ -70,6 +73,56 @@ func TestSentryToken_FromChunk(t *testing.T) {
 			},
 			wantErr: false,
 		},
+		{
+			name: "found, would be verified but for timeout",
+			s:    Scanner{client: common.SaneHttpClientTimeOut(1 * time.Microsecond)},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sentry super secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SentryToken,
+					Verified:     false,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: true,
+		},
+		{
+			name: "found and valid but unexpected api response",
+			s:    Scanner{client: common.ConstantResponseHttpClient(500, "")},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sentry super secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SentryToken,
+					Verified:     false,
+				},
+			},
+			wantErr:             false,
+			wantVerificationErr: true,
+		},
+		{
+			name: "found, good key but wrong scope",
+			s:    Scanner{client: common.ConstantResponseHttpClient(403, responseBody403)},
+			args: args{
+				ctx:    context.Background(),
+				data:   []byte(fmt.Sprintf("You can find a sentry super secret %s within", secret)),
+				verify: true,
+			},
+			want: []detectors.Result{
+				{
+					DetectorType: detectorspb.DetectorType_SentryToken,
+					Verified:     true,
+				},
+			},
+			wantErr: false,
+		},
 		{
 			name: "not found",
 			s:    Scanner{},
@@ -82,27 +135,46 @@ func TestSentryToken_FromChunk(t *testing.T) {
 			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)
+			got, err := tt.s.FromData(tt.args.ctx, tt.args.verify, tt.args.data)
 			if (err != nil) != tt.wantErr {
-				t.Errorf("SentryToken.FromData() error = %v, wantErr %v", err, tt.wantErr)
+				t.Errorf("Gitlab.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])
+					t.Fatal("no raw secret present")
+				}
+				if (got[i].VerificationError != nil) != tt.wantVerificationErr {
+					t.Fatalf("wantVerificationError = %v, verification error = %v,", tt.wantVerificationErr, got[i].VerificationError)
 				}
-				got[i].Raw = nil
 			}
-			if diff := pretty.Compare(got, tt.want); diff != "" {
-				t.Errorf("SentryToken.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
+			opts := cmpopts.IgnoreFields(detectors.Result{}, "Raw", "VerificationError")
+			if diff := cmp.Diff(got, tt.want, opts); diff != "" {
+				t.Errorf("Gitlab.FromData() %s diff: (-got +want)\n%s", tt.name, diff)
 			}
 		})
 	}
 }
 
+const responseBody403 = `
+[
+  {
+    "organization": {
+      "id": "911964",
+      "slug": "wigslap",
+      "status": {
+        "id": "active",
+        "name": "active"
+      },
+      "name": "wigslap"
+    }
+  }
+]
+`
+
 func BenchmarkFromData(benchmark *testing.B) {
 	ctx := context.Background()
 	s := Scanner{}

pkg/detectors/sourcegraph/sourcegraph.go

@@ -22,13 +22,13 @@ var _ detectors.Detector = (*Scanner)(nil)
 var (
 	defaultClient = common.SaneHttpClient()
 	// Make sure that your group is surrounded in boundary characters such as below to reduce false positives.
-	keyPat = regexp.MustCompile(detectors.PrefixRegex([]string{"sourcegraph", "sg_token"}) + `\b(sgp_[a-z0-9]{40})\b`)
+	keyPat = regexp.MustCompile(`\b(sgp_[a-f0-9]{40})\b`)
 )
 
 // 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{"sourcegraph", "sg_token"}
+	return []string{"sgp_"}
 }
 
 // FromData will find and optionally verify Sourcegraph secrets in a given set of bytes.

pkg/giturl/giturl.go

@@ -132,9 +132,9 @@ func GenerateLink(repo, commit, file string, line int64) string {
 		return baseLink
 
 	case providerAzure:
-		baseLink := repo + "?path=" + file + "&version=GB" + commit
+		baseLink := repo + "/commit/" + commit + "/" + file
 		if line > 0 {
-			baseLink += "&line=" + strconv.FormatInt(line, 10)
+			baseLink += "?line=" + strconv.FormatInt(line, 10)
 		}
 		return baseLink
 

pkg/giturl/giturl_test.go

@@ -148,7 +148,7 @@ func TestGenerateLink(t *testing.T) {
 				commit: "abcdef",
 				file:   "main.go",
 			},
-			want: "https://dev.azure.com/org/project/_git/repo?path=main.go&version=GBabcdef",
+			want: "https://dev.azure.com/org/project/_git/repo/commit/abcdef/main.go",
 		},
 		{
 			name: "Azure link gen with line",
@@ -158,7 +158,7 @@ func TestGenerateLink(t *testing.T) {
 				file:   "main.go",
 				line:   int64(20),
 			},
-			want: "https://dev.azure.com/org/project/_git/repo?path=main.go&version=GBabcdef&line=20",
+			want: "https://dev.azure.com/org/project/_git/repo/commit/abcdef/main.go?line=20",
 		},
 	}
 	for _, tt := range tests {