feat(15-03): add Elasticsearch, Kibana, and Splunk ReconSource implementations

- ElasticsearchSource: POST _search API with query_string, parse hits._source
- KibanaSource: GET saved_objects/_find API with kbn-xsrf header
- SplunkSource: GET search/jobs/export API with newline-delimited JSON parsing
- All sources use ciLogKeyPattern for key detection
- Tests use httptest mocks for each API endpoint

pkg/recon/sources/elasticsearch.go

@@ -0,0 +1,118 @@
+package sources
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"strings"
+	"time"
+
+	"golang.org/x/time/rate"
+
+	"github.com/salvacybersec/keyhunter/pkg/providers"
+	"github.com/salvacybersec/keyhunter/pkg/recon"
+)
+
+// ElasticsearchSource searches exposed Elasticsearch instances for documents
+// containing API keys. Many ES deployments are left unauthenticated on the
+// internet, allowing full-text search across all indexed data.
+type ElasticsearchSource struct {
+	BaseURL  string
+	Registry *providers.Registry
+	Limiters *recon.LimiterRegistry
+	Client   *Client
+}
+
+var _ recon.ReconSource = (*ElasticsearchSource)(nil)
+
+func (s *ElasticsearchSource) Name() string              { return "elasticsearch" }
+func (s *ElasticsearchSource) RateLimit() rate.Limit      { return rate.Every(2 * time.Second) }
+func (s *ElasticsearchSource) Burst() int                 { return 3 }
+func (s *ElasticsearchSource) RespectsRobots() bool       { return false }
+func (s *ElasticsearchSource) Enabled(_ recon.Config) bool { return true }
+
+// esSearchResponse represents the Elasticsearch _search response envelope.
+type esSearchResponse struct {
+	Hits struct {
+		Hits []esHit `json:"hits"`
+	} `json:"hits"`
+}
+
+type esHit struct {
+	Index  string          `json:"_index"`
+	ID     string          `json:"_id"`
+	Source json.RawMessage `json:"_source"`
+}
+
+func (s *ElasticsearchSource) Sweep(ctx context.Context, query string, out chan<- recon.Finding) error {
+	base := s.BaseURL
+	if base == "" {
+		base = "http://localhost:9200"
+	}
+	// If no explicit target was provided (still default) and query is not a URL, skip.
+	if base == "http://localhost:9200" && query != "" && !strings.HasPrefix(query, "http") {
+		return nil
+	}
+	client := s.Client
+	if client == nil {
+		client = NewClient()
+	}
+
+	queries := BuildQueries(s.Registry, "elasticsearch")
+	if len(queries) == 0 {
+		return nil
+	}
+
+	for _, q := range queries {
+		if err := ctx.Err(); err != nil {
+			return err
+		}
+
+		if s.Limiters != nil {
+			if err := s.Limiters.Wait(ctx, s.Name(), s.RateLimit(), s.Burst(), false); err != nil {
+				return err
+			}
+		}
+
+		searchURL := fmt.Sprintf("%s/_search", base)
+		body := fmt.Sprintf(`{"query":{"query_string":{"query":"%s"}},"size":20}`, q)
+		req, err := http.NewRequestWithContext(ctx, http.MethodPost, searchURL, bytes.NewBufferString(body))
+		if err != nil {
+			continue
+		}
+		req.Header.Set("Content-Type", "application/json")
+
+		resp, err := client.Do(ctx, req)
+		if err != nil {
+			continue
+		}
+
+		data, err := io.ReadAll(io.LimitReader(resp.Body, 512*1024))
+		_ = resp.Body.Close()
+		if err != nil {
+			continue
+		}
+
+		var result esSearchResponse
+		if err := json.Unmarshal(data, &result); err != nil {
+			continue
+		}
+
+		for _, hit := range result.Hits.Hits {
+			src := string(hit.Source)
+			if ciLogKeyPattern.MatchString(src) {
+				out <- recon.Finding{
+					ProviderName: q,
+					Source:       fmt.Sprintf("%s/%s/%s", base, hit.Index, hit.ID),
+					SourceType:   "recon:elasticsearch",
+					Confidence:   "medium",
+					DetectedAt:   time.Now(),
+				}
+			}
+		}
+	}
+	return nil
+}

pkg/recon/sources/elasticsearch_test.go

@@ -0,0 +1,120 @@
+package sources
+
+import (
+	"context"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+	"time"
+
+	"github.com/salvacybersec/keyhunter/pkg/providers"
+	"github.com/salvacybersec/keyhunter/pkg/recon"
+)
+
+func TestElasticsearch_Name(t *testing.T) {
+	s := &ElasticsearchSource{}
+	if s.Name() != "elasticsearch" {
+		t.Fatalf("expected elasticsearch, got %s", s.Name())
+	}
+}
+
+func TestElasticsearch_Enabled(t *testing.T) {
+	s := &ElasticsearchSource{}
+	if !s.Enabled(recon.Config{}) {
+		t.Fatal("ElasticsearchSource should always be enabled")
+	}
+}
+
+func TestElasticsearch_Sweep(t *testing.T) {
+	mux := http.NewServeMux()
+	mux.HandleFunc("/_search", func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{
+			"hits": {
+				"hits": [
+					{
+						"_index": "logs",
+						"_id": "abc123",
+						"_source": {
+							"message": "api_key = sk-proj-ABCDEF1234567890abcdef",
+							"level": "error"
+						}
+					}
+				]
+			}
+		}`))
+	})
+
+	srv := httptest.NewServer(mux)
+	defer srv.Close()
+
+	reg := providers.NewRegistryFromProviders([]providers.Provider{
+		{Name: "openai", Keywords: []string{"sk-proj-"}},
+	})
+
+	s := &ElasticsearchSource{
+		BaseURL:  srv.URL,
+		Registry: reg,
+		Client:   NewClient(),
+	}
+
+	out := make(chan recon.Finding, 10)
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	err := s.Sweep(ctx, "", out)
+	close(out)
+	if err != nil {
+		t.Fatalf("Sweep error: %v", err)
+	}
+
+	var findings []recon.Finding
+	for f := range out {
+		findings = append(findings, f)
+	}
+	if len(findings) == 0 {
+		t.Fatal("expected at least one finding from Elasticsearch")
+	}
+	if findings[0].SourceType != "recon:elasticsearch" {
+		t.Fatalf("expected recon:elasticsearch, got %s", findings[0].SourceType)
+	}
+}
+
+func TestElasticsearch_Sweep_NoHits(t *testing.T) {
+	mux := http.NewServeMux()
+	mux.HandleFunc("/_search", func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{"hits":{"hits":[]}}`))
+	})
+
+	srv := httptest.NewServer(mux)
+	defer srv.Close()
+
+	reg := providers.NewRegistryFromProviders([]providers.Provider{
+		{Name: "openai", Keywords: []string{"sk-proj-"}},
+	})
+
+	s := &ElasticsearchSource{
+		BaseURL:  srv.URL,
+		Registry: reg,
+		Client:   NewClient(),
+	}
+
+	out := make(chan recon.Finding, 10)
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	err := s.Sweep(ctx, "", out)
+	close(out)
+	if err != nil {
+		t.Fatalf("Sweep error: %v", err)
+	}
+
+	var findings []recon.Finding
+	for f := range out {
+		findings = append(findings, f)
+	}
+	if len(findings) != 0 {
+		t.Fatalf("expected no findings, got %d", len(findings))
+	}
+}

pkg/recon/sources/kibana.go

@@ -0,0 +1,114 @@
+package sources
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"time"
+
+	"golang.org/x/time/rate"
+
+	"github.com/salvacybersec/keyhunter/pkg/providers"
+	"github.com/salvacybersec/keyhunter/pkg/recon"
+)
+
+// KibanaSource searches exposed Kibana instances for API keys in saved objects
+// such as dashboards, visualizations, and index patterns. Many Kibana instances
+// are left unauthenticated, exposing the saved objects API.
+type KibanaSource struct {
+	BaseURL  string
+	Registry *providers.Registry
+	Limiters *recon.LimiterRegistry
+	Client   *Client
+}
+
+var _ recon.ReconSource = (*KibanaSource)(nil)
+
+func (s *KibanaSource) Name() string              { return "kibana" }
+func (s *KibanaSource) RateLimit() rate.Limit      { return rate.Every(2 * time.Second) }
+func (s *KibanaSource) Burst() int                 { return 3 }
+func (s *KibanaSource) RespectsRobots() bool       { return false }
+func (s *KibanaSource) Enabled(_ recon.Config) bool { return true }
+
+// kibanaSavedObjectsResponse represents the Kibana saved objects API response.
+type kibanaSavedObjectsResponse struct {
+	SavedObjects []kibanaSavedObject `json:"saved_objects"`
+}
+
+type kibanaSavedObject struct {
+	ID         string          `json:"id"`
+	Type       string          `json:"type"`
+	Attributes json.RawMessage `json:"attributes"`
+}
+
+func (s *KibanaSource) Sweep(ctx context.Context, query string, out chan<- recon.Finding) error {
+	base := s.BaseURL
+	if base == "" {
+		base = "http://localhost:5601"
+	}
+	client := s.Client
+	if client == nil {
+		client = NewClient()
+	}
+
+	queries := BuildQueries(s.Registry, "kibana")
+	if len(queries) == 0 {
+		return nil
+	}
+
+	for _, q := range queries {
+		if err := ctx.Err(); err != nil {
+			return err
+		}
+
+		if s.Limiters != nil {
+			if err := s.Limiters.Wait(ctx, s.Name(), s.RateLimit(), s.Burst(), false); err != nil {
+				return err
+			}
+		}
+
+		// Search saved objects (dashboards and visualizations).
+		searchURL := fmt.Sprintf(
+			"%s/api/saved_objects/_find?type=visualization&type=dashboard&search=%s&per_page=20",
+			base, url.QueryEscape(q),
+		)
+		req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
+		if err != nil {
+			continue
+		}
+		req.Header.Set("kbn-xsrf", "true")
+
+		resp, err := client.Do(ctx, req)
+		if err != nil {
+			continue
+		}
+
+		data, err := io.ReadAll(io.LimitReader(resp.Body, 512*1024))
+		_ = resp.Body.Close()
+		if err != nil {
+			continue
+		}
+
+		var result kibanaSavedObjectsResponse
+		if err := json.Unmarshal(data, &result); err != nil {
+			continue
+		}
+
+		for _, obj := range result.SavedObjects {
+			attrs := string(obj.Attributes)
+			if ciLogKeyPattern.MatchString(attrs) {
+				out <- recon.Finding{
+					ProviderName: q,
+					Source:       fmt.Sprintf("%s/app/kibana#/%s/%s", base, obj.Type, obj.ID),
+					SourceType:   "recon:kibana",
+					Confidence:   "medium",
+					DetectedAt:   time.Now(),
+				}
+			}
+		}
+	}
+	return nil
+}

pkg/recon/sources/kibana_test.go

@@ -0,0 +1,123 @@
+package sources
+
+import (
+	"context"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+	"time"
+
+	"github.com/salvacybersec/keyhunter/pkg/providers"
+	"github.com/salvacybersec/keyhunter/pkg/recon"
+)
+
+func TestKibana_Name(t *testing.T) {
+	s := &KibanaSource{}
+	if s.Name() != "kibana" {
+		t.Fatalf("expected kibana, got %s", s.Name())
+	}
+}
+
+func TestKibana_Enabled(t *testing.T) {
+	s := &KibanaSource{}
+	if !s.Enabled(recon.Config{}) {
+		t.Fatal("KibanaSource should always be enabled")
+	}
+}
+
+func TestKibana_Sweep(t *testing.T) {
+	mux := http.NewServeMux()
+	mux.HandleFunc("/api/saved_objects/_find", func(w http.ResponseWriter, r *http.Request) {
+		// Verify kbn-xsrf header is present.
+		if r.Header.Get("kbn-xsrf") == "" {
+			http.Error(w, "missing kbn-xsrf", http.StatusBadRequest)
+			return
+		}
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{
+			"saved_objects": [
+				{
+					"id": "vis-001",
+					"type": "visualization",
+					"attributes": {
+						"title": "API Usage",
+						"config": "api_key = sk-proj-ABCDEF1234567890abcdef"
+					}
+				}
+			]
+		}`))
+	})
+
+	srv := httptest.NewServer(mux)
+	defer srv.Close()
+
+	reg := providers.NewRegistryFromProviders([]providers.Provider{
+		{Name: "openai", Keywords: []string{"sk-proj-"}},
+	})
+
+	s := &KibanaSource{
+		BaseURL:  srv.URL,
+		Registry: reg,
+		Client:   NewClient(),
+	}
+
+	out := make(chan recon.Finding, 10)
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	err := s.Sweep(ctx, "", out)
+	close(out)
+	if err != nil {
+		t.Fatalf("Sweep error: %v", err)
+	}
+
+	var findings []recon.Finding
+	for f := range out {
+		findings = append(findings, f)
+	}
+	if len(findings) == 0 {
+		t.Fatal("expected at least one finding from Kibana")
+	}
+	if findings[0].SourceType != "recon:kibana" {
+		t.Fatalf("expected recon:kibana, got %s", findings[0].SourceType)
+	}
+}
+
+func TestKibana_Sweep_NoFindings(t *testing.T) {
+	mux := http.NewServeMux()
+	mux.HandleFunc("/api/saved_objects/_find", func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(`{"saved_objects":[]}`))
+	})
+
+	srv := httptest.NewServer(mux)
+	defer srv.Close()
+
+	reg := providers.NewRegistryFromProviders([]providers.Provider{
+		{Name: "openai", Keywords: []string{"sk-proj-"}},
+	})
+
+	s := &KibanaSource{
+		BaseURL:  srv.URL,
+		Registry: reg,
+		Client:   NewClient(),
+	}
+
+	out := make(chan recon.Finding, 10)
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	err := s.Sweep(ctx, "", out)
+	close(out)
+	if err != nil {
+		t.Fatalf("Sweep error: %v", err)
+	}
+
+	var findings []recon.Finding
+	for f := range out {
+		findings = append(findings, f)
+	}
+	if len(findings) != 0 {
+		t.Fatalf("expected no findings, got %d", len(findings))
+	}
+}

pkg/recon/sources/splunk.go

@@ -0,0 +1,122 @@
+package sources
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"strings"
+	"time"
+
+	"golang.org/x/time/rate"
+
+	"github.com/salvacybersec/keyhunter/pkg/providers"
+	"github.com/salvacybersec/keyhunter/pkg/recon"
+)
+
+// SplunkSource searches exposed Splunk instances for API keys in log data.
+// Exposed Splunk Web interfaces may allow unauthenticated search via the
+// REST API, especially in development or misconfigured environments.
+type SplunkSource struct {
+	BaseURL  string
+	Registry *providers.Registry
+	Limiters *recon.LimiterRegistry
+	Client   *Client
+}
+
+var _ recon.ReconSource = (*SplunkSource)(nil)
+
+func (s *SplunkSource) Name() string              { return "splunk" }
+func (s *SplunkSource) RateLimit() rate.Limit      { return rate.Every(3 * time.Second) }
+func (s *SplunkSource) Burst() int                 { return 2 }
+func (s *SplunkSource) RespectsRobots() bool       { return false }
+func (s *SplunkSource) Enabled(_ recon.Config) bool { return true }
+
+// splunkResult represents a single result row from Splunk search export.
+type splunkResult struct {
+	Result json.RawMessage `json:"result"`
+	Raw    string          `json:"_raw"`
+}
+
+func (s *SplunkSource) Sweep(ctx context.Context, query string, out chan<- recon.Finding) error {
+	base := s.BaseURL
+	if base == "" {
+		base = "https://localhost:8089"
+	}
+	// If no explicit target was provided (still default) and query is not a URL, skip.
+	if base == "https://localhost:8089" && query != "" && !strings.HasPrefix(query, "http") {
+		return nil
+	}
+	client := s.Client
+	if client == nil {
+		client = NewClient()
+	}
+
+	queries := BuildQueries(s.Registry, "splunk")
+	if len(queries) == 0 {
+		return nil
+	}
+
+	for _, q := range queries {
+		if err := ctx.Err(); err != nil {
+			return err
+		}
+
+		if s.Limiters != nil {
+			if err := s.Limiters.Wait(ctx, s.Name(), s.RateLimit(), s.Burst(), false); err != nil {
+				return err
+			}
+		}
+
+		searchURL := fmt.Sprintf(
+			"%s/services/search/jobs/export?search=%s&output_mode=json&count=20",
+			base, url.QueryEscape("search "+q),
+		)
+		req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
+		if err != nil {
+			continue
+		}
+
+		resp, err := client.Do(ctx, req)
+		if err != nil {
+			continue
+		}
+
+		data, err := io.ReadAll(io.LimitReader(resp.Body, 512*1024))
+		_ = resp.Body.Close()
+		if err != nil {
+			continue
+		}
+
+		// Splunk export returns newline-delimited JSON objects.
+		for _, line := range strings.Split(string(data), "\n") {
+			line = strings.TrimSpace(line)
+			if line == "" {
+				continue
+			}
+
+			var sr splunkResult
+			if err := json.Unmarshal([]byte(line), &sr); err != nil {
+				continue
+			}
+
+			content := sr.Raw
+			if content == "" {
+				content = string(sr.Result)
+			}
+
+			if ciLogKeyPattern.MatchString(content) {
+				out <- recon.Finding{
+					ProviderName: q,
+					Source:       fmt.Sprintf("%s/services/search/jobs/export", base),
+					SourceType:   "recon:splunk",
+					Confidence:   "medium",
+					DetectedAt:   time.Now(),
+				}
+			}
+		}
+	}
+	return nil
+}

pkg/recon/sources/splunk_test.go

@@ -0,0 +1,110 @@
+package sources
+
+import (
+	"context"
+	"net/http"
+	"net/http/httptest"
+	"testing"
+	"time"
+
+	"github.com/salvacybersec/keyhunter/pkg/providers"
+	"github.com/salvacybersec/keyhunter/pkg/recon"
+)
+
+func TestSplunk_Name(t *testing.T) {
+	s := &SplunkSource{}
+	if s.Name() != "splunk" {
+		t.Fatalf("expected splunk, got %s", s.Name())
+	}
+}
+
+func TestSplunk_Enabled(t *testing.T) {
+	s := &SplunkSource{}
+	if !s.Enabled(recon.Config{}) {
+		t.Fatal("SplunkSource should always be enabled")
+	}
+}
+
+func TestSplunk_Sweep(t *testing.T) {
+	mux := http.NewServeMux()
+	mux.HandleFunc("/services/search/jobs/export", func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "application/json")
+		// Splunk returns newline-delimited JSON.
+		_, _ = w.Write([]byte(`{"result":{"_raw":"Setting secret_key = sk-proj-ABCDEF1234567890abcdef"},"_raw":"Setting secret_key = sk-proj-ABCDEF1234567890abcdef"}
+{"result":{"_raw":"normal log line no keys here"},"_raw":"normal log line no keys here"}
+`))
+	})
+
+	srv := httptest.NewServer(mux)
+	defer srv.Close()
+
+	reg := providers.NewRegistryFromProviders([]providers.Provider{
+		{Name: "openai", Keywords: []string{"sk-proj-"}},
+	})
+
+	s := &SplunkSource{
+		BaseURL:  srv.URL,
+		Registry: reg,
+		Client:   NewClient(),
+	}
+
+	out := make(chan recon.Finding, 10)
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	err := s.Sweep(ctx, "", out)
+	close(out)
+	if err != nil {
+		t.Fatalf("Sweep error: %v", err)
+	}
+
+	var findings []recon.Finding
+	for f := range out {
+		findings = append(findings, f)
+	}
+	if len(findings) == 0 {
+		t.Fatal("expected at least one finding from Splunk")
+	}
+	if findings[0].SourceType != "recon:splunk" {
+		t.Fatalf("expected recon:splunk, got %s", findings[0].SourceType)
+	}
+}
+
+func TestSplunk_Sweep_NoResults(t *testing.T) {
+	mux := http.NewServeMux()
+	mux.HandleFunc("/services/search/jobs/export", func(w http.ResponseWriter, r *http.Request) {
+		w.Header().Set("Content-Type", "application/json")
+		_, _ = w.Write([]byte(``))
+	})
+
+	srv := httptest.NewServer(mux)
+	defer srv.Close()
+
+	reg := providers.NewRegistryFromProviders([]providers.Provider{
+		{Name: "openai", Keywords: []string{"sk-proj-"}},
+	})
+
+	s := &SplunkSource{
+		BaseURL:  srv.URL,
+		Registry: reg,
+		Client:   NewClient(),
+	}
+
+	out := make(chan recon.Finding, 10)
+	ctx, cancel := context.WithTimeout(context.Background(), 5*time.Second)
+	defer cancel()
+
+	err := s.Sweep(ctx, "", out)
+	close(out)
+	if err != nil {
+		t.Fatalf("Sweep error: %v", err)
+	}
+
+	var findings []recon.Finding
+	for f := range out {
+		findings = append(findings, f)
+	}
+	if len(findings) != 0 {
+		t.Fatalf("expected no findings, got %d", len(findings))
+	}
+}