keyhunter/pkg/recon/sources/notion.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"
)

// NotionSource searches publicly shared Notion pages for leaked API keys.
// Notion pages shared with "anyone with the link" are indexable by search
// engines. This source uses a dorking approach to discover such pages and
// then scrapes their content for credentials.
type NotionSource struct {
	BaseURL  string
	Registry *providers.Registry
	Limiters *recon.LimiterRegistry
	Client   *Client
}

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

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

// notionSearchResponse represents dork search results pointing to Notion pages.
type notionSearchResponse struct {
	Results []notionSearchResult `json:"results"`
}

type notionSearchResult struct {
	URL   string `json:"url"`
	Title string `json:"title"`
}

func (s *NotionSource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
	base := s.BaseURL
	if base == "" {
		base = "https://search.notion.dev"
	}
	client := s.Client
	if client == nil {
		client = NewClient()
	}

	queries := BuildQueries(s.Registry, "notion")
	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 for public Notion pages via dorking.
		searchURL := fmt.Sprintf("%s/search?q=%s&format=json",
			base, url.QueryEscape("site:notion.site OR site:notion.so "+q))
		req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
		if err != nil {
			continue
		}
		req.Header.Set("Accept", "application/json")

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

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

		var results notionSearchResponse
		if err := json.Unmarshal(body, &results); err != nil {
			continue
		}

		// Fetch each discovered Notion page and scan for keys.
		for _, result := range results.Results {
			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
				}
			}

			pageReq, err := http.NewRequestWithContext(ctx, http.MethodGet, result.URL, nil)
			if err != nil {
				continue
			}

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

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

			if ciLogKeyPattern.Match(pageBody) {
				out <- recon.Finding{
					ProviderName: q,
					Source:       result.URL,
					SourceType:   "recon:notion",
					Confidence:   "medium",
					DetectedAt:   time.Now(),
				}
			}
		}
	}
	return nil
}