keyhunter/pkg/recon/sources/jenkins.go

package sources

import (
	"context"
	"encoding/json"
	"fmt"
<<<<<<< HEAD
	"net/http"
	"strings"
=======
	"io"
	"net/http"
>>>>>>> worktree-agent-adad8c10
	"time"

	"golang.org/x/time/rate"

	"github.com/salvacybersec/keyhunter/pkg/providers"
	"github.com/salvacybersec/keyhunter/pkg/recon"
)

<<<<<<< HEAD
// JenkinsSource searches publicly accessible Jenkins instances for build
// console output containing leaked API keys. It queries the Jenkins JSON API
// at /api/json to enumerate jobs and their latest builds. Credentialless --
// targets open Jenkins instances discovered via dorking or IoT scanners.
=======
// JenkinsSource scrapes publicly accessible Jenkins build consoles for leaked
// API keys. Many Jenkins instances are exposed to the internet without
// authentication, and build console output frequently contains printed
// environment variables or secrets passed via command-line arguments.
>>>>>>> worktree-agent-adad8c10
type JenkinsSource struct {
	BaseURL  string
	Registry *providers.Registry
	Limiters *recon.LimiterRegistry
	Client   *Client
}

var _ recon.ReconSource = (*JenkinsSource)(nil)

<<<<<<< HEAD
func (s *JenkinsSource) Name() string             { return "jenkins" }
func (s *JenkinsSource) RateLimit() rate.Limit    { return rate.Every(3 * time.Second) }
func (s *JenkinsSource) Burst() int               { return 1 }
func (s *JenkinsSource) RespectsRobots() bool     { return true }
func (s *JenkinsSource) Enabled(_ recon.Config) bool { return true }

=======
func (s *JenkinsSource) Name() string              { return "jenkins" }
func (s *JenkinsSource) RateLimit() rate.Limit      { return rate.Every(3 * time.Second) }
func (s *JenkinsSource) Burst() int                 { return 2 }
func (s *JenkinsSource) RespectsRobots() bool       { return true }
func (s *JenkinsSource) Enabled(_ recon.Config) bool { return true }

// jenkinsJobsResponse represents the Jenkins API jobs listing.
>>>>>>> worktree-agent-adad8c10
type jenkinsJobsResponse struct {
	Jobs []jenkinsJob `json:"jobs"`
}

type jenkinsJob struct {
<<<<<<< HEAD
	Name     string       `json:"name"`
	URL      string       `json:"url"`
	LastBuild *jenkinsBuild `json:"lastBuild"`
}

type jenkinsBuild struct {
	Number int    `json:"number"`
	URL    string `json:"url"`
=======
	Name  string `json:"name"`
	URL   string `json:"url"`
	Color string `json:"color"`
>>>>>>> worktree-agent-adad8c10
}

func (s *JenkinsSource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
	base := s.BaseURL
	if base == "" {
<<<<<<< HEAD
		base = "https://jenkins.example.com"
=======
		return nil // No default; Jenkins instances are discovered via dorking
>>>>>>> worktree-agent-adad8c10
	}
	client := s.Client
	if client == nil {
		client = NewClient()
	}

	queries := BuildQueries(s.Registry, "jenkins")
<<<<<<< HEAD
	kwIndex := jenkinsKeywordIndex(s.Registry)
=======
	if len(queries) == 0 {
		return nil
	}
>>>>>>> worktree-agent-adad8c10

	for _, q := range queries {
		if err := ctx.Err(); err != nil {
			return err
		}
<<<<<<< HEAD
=======

>>>>>>> worktree-agent-adad8c10
		if s.Limiters != nil {
			if err := s.Limiters.Wait(ctx, s.Name(), s.RateLimit(), s.Burst(), false); err != nil {
				return err
			}
		}

<<<<<<< HEAD
		endpoint := fmt.Sprintf("%s/api/json?tree=jobs[name,url,lastBuild[number,url]]", base)
		req, err := http.NewRequestWithContext(ctx, http.MethodGet, endpoint, nil)
		if err != nil {
			return fmt.Errorf("jenkins: build request: %w", err)
=======
		// List jobs from the Jenkins API.
		jobsURL := fmt.Sprintf("%s/api/json?tree=jobs[name,url,color]", base)
		req, err := http.NewRequestWithContext(ctx, http.MethodGet, jobsURL, nil)
		if err != nil {
			continue
>>>>>>> worktree-agent-adad8c10
		}
		req.Header.Set("Accept", "application/json")

		resp, err := client.Do(ctx, req)
		if err != nil {
			continue
		}

<<<<<<< HEAD
		var result jenkinsJobsResponse
		decErr := json.NewDecoder(resp.Body).Decode(&result)
		_ = resp.Body.Close()
		if decErr != nil {
			continue
		}

		provName := kwIndex[strings.ToLower(q)]
		for _, job := range result.Jobs {
			if job.LastBuild == nil {
				continue
			}
			source := job.LastBuild.URL
			if source == "" {
				source = fmt.Sprintf("%s/job/%s/%d/console", base, job.Name, job.LastBuild.Number)
			}
			f := recon.Finding{
				ProviderName: provName,
				Confidence:   "low",
				Source:       source,
				SourceType:   "recon:jenkins",
				DetectedAt:   time.Now(),
			}
			select {
			case out <- f:
			case <-ctx.Done():
				return ctx.Err()
=======
		var jobs jenkinsJobsResponse
		if err := json.NewDecoder(resp.Body).Decode(&jobs); err != nil {
			_ = resp.Body.Close()
			continue
		}
		_ = resp.Body.Close()

		for _, job := range jobs.Jobs {
			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
				}
			}

			// Fetch the last build console output.
			consoleURL := fmt.Sprintf("%s/job/%s/lastBuild/consoleText", base, job.Name)
			consoleReq, err := http.NewRequestWithContext(ctx, http.MethodGet, consoleURL, nil)
			if err != nil {
				continue
			}

			consoleResp, err := client.Do(ctx, consoleReq)
			if err != nil {
				continue
			}

			body, err := io.ReadAll(io.LimitReader(consoleResp.Body, 256*1024))
			_ = consoleResp.Body.Close()
			if err != nil {
				continue
			}

			if ciLogKeyPattern.Match(body) {
				out <- recon.Finding{
					ProviderName: q,
					Source:       consoleURL,
					SourceType:   "recon:jenkins",
					Confidence:   "medium",
					DetectedAt:   time.Now(),
				}
>>>>>>> worktree-agent-adad8c10
			}
		}
	}
	return nil
}
<<<<<<< HEAD

func jenkinsKeywordIndex(reg *providers.Registry) map[string]string {
	m := make(map[string]string)
	if reg == nil {
		return m
	}
	for _, p := range reg.List() {
		for _, k := range p.Keywords {
			kl := strings.ToLower(strings.TrimSpace(k))
			if kl != "" {
				if _, exists := m[kl]; !exists {
					m[kl] = p.Name
				}
			}
		}
	}
	return m
}
=======
>>>>>>> worktree-agent-adad8c10