4b268d109f
- NpmSource searches npm registry JSON API for provider keywords - PyPISource scrapes pypi.org search HTML for project links - Both credentialless, rate-limited at 1 req/2s, burst 2 - httptest-based tests verify Sweep, ctx cancellation, Name/Rate/Burst
115 lines
2.6 KiB
Go
115 lines
2.6 KiB
Go
package sources
|
|
|
|
import (
|
|
"context"
|
|
"encoding/json"
|
|
"fmt"
|
|
"net/http"
|
|
"net/url"
|
|
"time"
|
|
|
|
"golang.org/x/time/rate"
|
|
|
|
"github.com/salvacybersec/keyhunter/pkg/providers"
|
|
"github.com/salvacybersec/keyhunter/pkg/recon"
|
|
)
|
|
|
|
// NpmSource searches the npm registry for packages matching provider keywords.
|
|
// No credentials required. Emits findings tagged SourceType=recon:npm.
|
|
type NpmSource struct {
|
|
BaseURL string
|
|
Registry *providers.Registry
|
|
Limiters *recon.LimiterRegistry
|
|
Client *Client
|
|
}
|
|
|
|
var _ recon.ReconSource = (*NpmSource)(nil)
|
|
|
|
func (s *NpmSource) Name() string { return "npm" }
|
|
func (s *NpmSource) RateLimit() rate.Limit { return rate.Every(2 * time.Second) }
|
|
func (s *NpmSource) Burst() int { return 2 }
|
|
func (s *NpmSource) RespectsRobots() bool { return false }
|
|
func (s *NpmSource) Enabled(_ recon.Config) bool { return true }
|
|
|
|
// npm search JSON response structs.
|
|
type npmSearchResponse struct {
|
|
Objects []npmObject `json:"objects"`
|
|
}
|
|
|
|
type npmObject struct {
|
|
Package npmPackage `json:"package"`
|
|
}
|
|
|
|
type npmPackage struct {
|
|
Name string `json:"name"`
|
|
Links npmLinks `json:"links"`
|
|
}
|
|
|
|
type npmLinks struct {
|
|
Npm string `json:"npm"`
|
|
}
|
|
|
|
func (s *NpmSource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
|
|
base := s.BaseURL
|
|
if base == "" {
|
|
base = "https://registry.npmjs.org"
|
|
}
|
|
client := s.Client
|
|
if client == nil {
|
|
client = NewClient()
|
|
}
|
|
|
|
queries := BuildQueries(s.Registry, "npm")
|
|
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
|
|
}
|
|
}
|
|
|
|
searchURL := fmt.Sprintf("%s/-/v1/search?text=%s&size=20", base, url.QueryEscape(q))
|
|
req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
|
|
if err != nil {
|
|
return fmt.Errorf("npm: build req: %w", err)
|
|
}
|
|
|
|
resp, err := client.Do(ctx, req)
|
|
if err != nil {
|
|
return fmt.Errorf("npm: fetch: %w", err)
|
|
}
|
|
|
|
var result npmSearchResponse
|
|
if err := json.NewDecoder(resp.Body).Decode(&result); err != nil {
|
|
_ = resp.Body.Close()
|
|
return fmt.Errorf("npm: decode json: %w", err)
|
|
}
|
|
_ = resp.Body.Close()
|
|
|
|
for _, obj := range result.Objects {
|
|
if err := ctx.Err(); err != nil {
|
|
return err
|
|
}
|
|
source := obj.Package.Links.Npm
|
|
if source == "" {
|
|
source = fmt.Sprintf("https://www.npmjs.com/package/%s", obj.Package.Name)
|
|
}
|
|
out <- recon.Finding{
|
|
ProviderName: "",
|
|
Source: source,
|
|
SourceType: "recon:npm",
|
|
Confidence: "low",
|
|
DetectedAt: time.Now(),
|
|
}
|
|
}
|
|
}
|
|
return nil
|
|
}
|