keyhunter/pkg/recon/sources/sentry.go

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"
)

// SentrySource searches exposed Sentry instances for API keys in error reports.
// Self-hosted Sentry installations may have the API accessible without
// authentication, exposing error events that commonly contain API keys in
// request headers, environment variables, and stack traces.
type SentrySource struct {
	BaseURL  string
	Registry *providers.Registry
	Limiters *recon.LimiterRegistry
	Client   *Client
}

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

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

// sentryIssue represents a Sentry issue from the issues list API.
type sentryIssue struct {
	ID    string `json:"id"`
	Title string `json:"title"`
}

// sentryEvent represents a Sentry event from the events API.
type sentryEvent struct {
	EventID string          `json:"eventID"`
	Tags    json.RawMessage `json:"tags"`
	Context json.RawMessage `json:"context"`
	Entries json.RawMessage `json:"entries"`
}

func (s *SentrySource) Sweep(ctx context.Context, query string, out chan<- recon.Finding) error {
	base := s.BaseURL
	if base == "" {
		base = "https://sentry.example.com"
	}
	client := s.Client
	if client == nil {
		client = NewClient()
	}

	queries := BuildQueries(s.Registry, "sentry")
	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 issues matching keyword.
		issuesURL := fmt.Sprintf(
			"%s/api/0/issues/?query=%s&limit=10",
			base, url.QueryEscape(q),
		)
		req, err := http.NewRequestWithContext(ctx, http.MethodGet, issuesURL, 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
		}

		var issues []sentryIssue
		if err := json.Unmarshal(data, &issues); err != nil {
			continue
		}

		// Fetch events for each issue.
		for _, issue := range issues {
			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
				}
			}

			eventsURL := fmt.Sprintf("%s/api/0/issues/%s/events/?limit=5", base, issue.ID)
			evReq, err := http.NewRequestWithContext(ctx, http.MethodGet, eventsURL, nil)
			if err != nil {
				continue
			}

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

			evData, err := io.ReadAll(io.LimitReader(evResp.Body, 512*1024))
			_ = evResp.Body.Close()
			if err != nil {
				continue
			}

			var events []sentryEvent
			if err := json.Unmarshal(evData, &events); err != nil {
				continue
			}

			for _, ev := range events {
				content := string(ev.Tags) + string(ev.Context) + string(ev.Entries)
				if ciLogKeyPattern.MatchString(content) {
					out <- recon.Finding{
						ProviderName: q,
						Source:       fmt.Sprintf("%s/issues/%s/events/%s", base, issue.ID, ev.EventID),
						SourceType:   "recon:sentry",
						Confidence:   "medium",
						DetectedAt:   time.Now(),
					}
				}
			}
		}
	}
	return nil
}