merge: phase 14-04 register wiring

pkg/recon/sources/commoncrawl.go

@@ -1,12 +1,21 @@
 package sources
 
 import (
+<<<<<<< HEAD
 	"bufio"
 	"context"
 	"encoding/json"
 	"fmt"
 	"net/http"
 	"net/url"
+=======
+	"bytes"
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+>>>>>>> worktree-agent-adad8c10
 	"time"
 
 	"golang.org/x/time/rate"
@@ -15,6 +24,7 @@ import (
 	"github.com/salvacybersec/keyhunter/pkg/recon"
 )
 
+<<<<<<< HEAD
 // CommonCrawlSource implements recon.ReconSource against the CommonCrawl
 // Index Server API. It queries index.commoncrawl.org for pages matching
 // provider keywords in the CC index.
@@ -57,6 +67,41 @@ func (s *CommonCrawlSource) Sweep(ctx context.Context, _ string, out chan<- reco
 	idx := s.IndexName
 	if idx == "" {
 		idx = "CC-MAIN-2024-10"
+=======
+// CommonCrawlSource searches the Common Crawl index for web pages that may
+// contain leaked API keys. Common Crawl archives petabytes of web content;
+// its CDX API allows searching by URL pattern to find pages that historically
+// exposed secrets.
+type CommonCrawlSource struct {
+	BaseURL  string
+	Registry *providers.Registry
+	Limiters *recon.LimiterRegistry
+	Client   *Client
+}
+
+var _ recon.ReconSource = (*CommonCrawlSource)(nil)
+
+func (s *CommonCrawlSource) Name() string              { return "commoncrawl" }
+func (s *CommonCrawlSource) RateLimit() rate.Limit      { return rate.Every(5 * time.Second) }
+func (s *CommonCrawlSource) Burst() int                 { return 1 }
+func (s *CommonCrawlSource) RespectsRobots() bool       { return true }
+func (s *CommonCrawlSource) Enabled(_ recon.Config) bool { return true }
+
+// ccIndexResult represents a single Common Crawl CDX index record.
+type ccIndexResult struct {
+	URL       string `json:"url"`
+	Timestamp string `json:"timestamp"`
+	Status    string `json:"status"`
+	Filename  string `json:"filename"`
+	Length    string `json:"length"`
+	Offset    string `json:"offset"`
+}
+
+func (s *CommonCrawlSource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
+	base := s.BaseURL
+	if base == "" {
+		base = "https://index.commoncrawl.org/CC-MAIN-2024-10-index"
+>>>>>>> worktree-agent-adad8c10
 	}
 	client := s.Client
 	if client == nil {
@@ -79,17 +124,26 @@ func (s *CommonCrawlSource) Sweep(ctx context.Context, _ string, out chan<- reco
 			}
 		}
 
+<<<<<<< HEAD
 		// CC Index API: output=json returns NDJSON, limit=50 bounds the response.
 		endpoint := fmt.Sprintf("%s/%s-index?url=*&output=json&limit=50&filter=status:200&query=%s",
 			base, idx, url.QueryEscape(q))
 		req, err := http.NewRequestWithContext(ctx, http.MethodGet, endpoint, nil)
 		if err != nil {
 			return fmt.Errorf("commoncrawl: build req: %w", err)
+=======
+		// CDX API: search for URLs matching the query.
+		searchURL := fmt.Sprintf("%s?url=*%s*&output=json&limit=10", base, q)
+		req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
+		if err != nil {
+			continue
+>>>>>>> worktree-agent-adad8c10
 		}
 		req.Header.Set("Accept", "application/json")
 
 		resp, err := client.Do(ctx, req)
 		if err != nil {
+<<<<<<< HEAD
 			// Non-fatal: skip this keyword on transient errors.
 			continue
 		}
@@ -112,10 +166,43 @@ func (s *CommonCrawlSource) Sweep(ctx context.Context, _ string, out chan<- reco
 			f := recon.Finding{
 				ProviderName: "",
 				Source:       rec.URL,
+=======
+			continue
+		}
+
+		body, err := io.ReadAll(io.LimitReader(resp.Body, 128*1024))
+		_ = resp.Body.Close()
+		if err != nil {
+			continue
+		}
+
+		// Common Crawl returns NDJSON (newline-delimited JSON).
+		// Parse each line as a separate JSON object.
+		var results []ccIndexResult
+		dec := json.NewDecoder(bytes.NewReader(body))
+		for dec.More() {
+			var r ccIndexResult
+			if err := dec.Decode(&r); err != nil {
+				break
+			}
+			results = append(results, r)
+		}
+
+		for _, r := range results {
+			if err := ctx.Err(); err != nil {
+				return err
+			}
+
+			// Each indexed URL is a potential leak location; emit as finding.
+			out <- recon.Finding{
+				ProviderName: q,
+				Source:       r.URL,
+>>>>>>> worktree-agent-adad8c10
 				SourceType:   "recon:commoncrawl",
 				Confidence:   "low",
 				DetectedAt:   time.Now(),
 			}
+<<<<<<< HEAD
 			select {
 			case out <- f:
 			case <-ctx.Done():
@@ -136,3 +223,9 @@ type ccIndexRecord struct {
 	Timestamp string `json:"timestamp"`
 	Status    string `json:"status"`
 }
+=======
+		}
+	}
+	return nil
+}
+>>>>>>> worktree-agent-adad8c10