feat(14-04): implement 7 Phase 14 sources (CI/CD, archives, JS bundles)

- TravisCISource: scrapes public Travis CI build logs for API key leaks
- GitHubActionsSource: searches Actions workflow logs (requires GitHub token)
- CircleCISource: scrapes CircleCI pipeline logs (requires CircleCI token)
- JenkinsSource: scrapes public Jenkins console output for leaked secrets
- WaybackMachineSource: searches Wayback Machine CDX for archived key leaks
- CommonCrawlSource: searches Common Crawl index for exposed pages
- JSBundleSource: probes JS bundles for embedded API key literals

pkg/recon/sources/jsbundle.go

@@ -0,0 +1,116 @@
+package sources
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"net/http"
+	"regexp"
+	"time"
+
+	"golang.org/x/time/rate"
+
+	"github.com/salvacybersec/keyhunter/pkg/providers"
+	"github.com/salvacybersec/keyhunter/pkg/recon"
+)
+
+// JSBundleSource analyzes public JavaScript bundles for embedded API keys.
+// Modern build tools (Webpack, Vite, esbuild, Rollup) inline environment
+// variables and configuration at build time. This source probes common bundle
+// paths and scans the minified JS for API key patterns, complementing
+// WebpackSource by targeting raw key literals rather than env var prefixes.
+type JSBundleSource struct {
+	BaseURL  string
+	Registry *providers.Registry
+	Limiters *recon.LimiterRegistry
+	Client   *Client
+}
+
+var _ recon.ReconSource = (*JSBundleSource)(nil)
+
+func (s *JSBundleSource) Name() string              { return "jsbundle" }
+func (s *JSBundleSource) RateLimit() rate.Limit      { return rate.Every(3 * time.Second) }
+func (s *JSBundleSource) Burst() int                 { return 2 }
+func (s *JSBundleSource) RespectsRobots() bool       { return true }
+func (s *JSBundleSource) Enabled(_ recon.Config) bool { return true }
+
+// jsBundleKeyPattern matches literal API key assignments commonly found in
+// minified JS bundles (e.g., apiKey:"sk-proj-...", "Authorization":"Bearer sk-...").
+var jsBundleKeyPattern = regexp.MustCompile(`(?i)(?:api[_-]?key|secret|token|authorization|bearer)\s*[=:"']+\s*['"]?([a-zA-Z0-9_\-]{20,})['"]?`)
+
+// jsBundlePaths are common locations for production JS bundles.
+var jsBundlePaths = []string{
+	"/static/js/main.js",
+	"/static/js/app.js",
+	"/static/js/vendor.js",
+	"/dist/app.js",
+	"/dist/main.js",
+	"/assets/app.js",
+	"/assets/index.js",
+	"/js/app.js",
+	"/_next/static/chunks/main.js",
+	"/_next/static/chunks/pages/_app.js",
+}
+
+func (s *JSBundleSource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
+	base := s.BaseURL
+	if base == "" {
+		return nil
+	}
+	client := s.Client
+	if client == nil {
+		client = NewClient()
+	}
+
+	queries := BuildQueries(s.Registry, "jsbundle")
+	if len(queries) == 0 {
+		return nil
+	}
+
+	for _, q := range queries {
+		if err := ctx.Err(); err != nil {
+			return err
+		}
+
+		for _, path := range jsBundlePaths {
+			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
+				}
+			}
+
+			probeURL := fmt.Sprintf("%s%s", base, path)
+			req, err := http.NewRequestWithContext(ctx, http.MethodGet, probeURL, nil)
+			if err != nil {
+				continue
+			}
+
+			resp, err := client.Do(ctx, req)
+			if err != nil {
+				continue
+			}
+
+			body, err := io.ReadAll(io.LimitReader(resp.Body, 1024*1024)) // 1MB max for JS bundles
+			_ = resp.Body.Close()
+			if err != nil {
+				continue
+			}
+
+			if jsBundleKeyPattern.Match(body) {
+				out <- recon.Finding{
+					ProviderName: q,
+					Source:       probeURL,
+					SourceType:   "recon:jsbundle",
+					Confidence:   "medium",
+					DetectedAt:   time.Now(),
+				}
+				break // one finding per query is sufficient
+			}
+		}
+	}
+	return nil
+}