feat(10-07): add Replit and CodeSandbox scraping sources
- ReplitSource scrapes /search HTML extracting /@user/repl anchors - CodeSandboxSource scrapes /search HTML extracting /s/slug anchors - Both use golang.org/x/net/html parser, 10 req/min rate, RespectsRobots=true - 10 httptest-backed tests covering extraction, ctx cancel, rate/name assertions
This commit is contained in:
salvacybersec
95
pkg/recon/sources/codesandbox.go
Normal file
95
pkg/recon/sources/codesandbox.go
Normal file
@@ -0,0 +1,95 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"regexp"
|
||||
"time"
|
||||
|
||||
"golang.org/x/time/rate"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
// CodeSandboxSource scrapes codesandbox.io search results for references to
|
||||
// provider keywords. Mirrors ReplitSource: robots-respecting, 10 req/min,
|
||||
// no credentials required.
|
||||
type CodeSandboxSource struct {
|
||||
BaseURL string
|
||||
Registry *providers.Registry
|
||||
Limiters *recon.LimiterRegistry
|
||||
Client *Client
|
||||
}
|
||||
|
||||
// codeSandboxLinkRE matches /s/<slug> sandbox result links.
|
||||
var codeSandboxLinkRE = regexp.MustCompile(`^/s/[a-zA-Z0-9-]+$`)
|
||||
|
||||
// Compile-time assertion that CodeSandboxSource satisfies recon.ReconSource.
|
||||
var _ recon.ReconSource = (*CodeSandboxSource)(nil)
|
||||
|
||||
func (s *CodeSandboxSource) Name() string { return "codesandbox" }
|
||||
func (s *CodeSandboxSource) RateLimit() rate.Limit { return rate.Every(6 * time.Second) }
|
||||
func (s *CodeSandboxSource) Burst() int { return 1 }
|
||||
func (s *CodeSandboxSource) RespectsRobots() bool { return true }
|
||||
func (s *CodeSandboxSource) Enabled(_ recon.Config) bool { return true }
|
||||
|
||||
// Sweep runs a CodeSandbox search per provider keyword and emits one Finding
|
||||
// per matched result anchor.
|
||||
func (s *CodeSandboxSource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
|
||||
base := s.BaseURL
|
||||
if base == "" {
|
||||
base = "https://codesandbox.io"
|
||||
}
|
||||
client := s.Client
|
||||
if client == nil {
|
||||
client = NewClient()
|
||||
}
|
||||
|
||||
queries := BuildQueries(s.Registry, "codesandbox")
|
||||
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/search?query=%s&type=sandboxes", base, url.QueryEscape(q))
|
||||
req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
|
||||
if err != nil {
|
||||
return fmt.Errorf("codesandbox: build req: %w", err)
|
||||
}
|
||||
resp, err := client.Do(ctx, req)
|
||||
if err != nil {
|
||||
return fmt.Errorf("codesandbox: fetch: %w", err)
|
||||
}
|
||||
links, err := extractAnchorHrefs(resp.Body, codeSandboxLinkRE)
|
||||
_ = resp.Body.Close()
|
||||
if err != nil {
|
||||
return fmt.Errorf("codesandbox: parse html: %w", err)
|
||||
}
|
||||
|
||||
for _, href := range links {
|
||||
if err := ctx.Err(); err != nil {
|
||||
return err
|
||||
}
|
||||
out <- recon.Finding{
|
||||
Source: base + href,
|
||||
SourceType: "recon:codesandbox",
|
||||
Confidence: "low",
|
||||
DetectedAt: time.Now(),
|
||||
}
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
109
pkg/recon/sources/codesandbox_test.go
Normal file
109
pkg/recon/sources/codesandbox_test.go
Normal file
@@ -0,0 +1,109 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
func codesandboxTestRegistry() *providers.Registry {
|
||||
return providers.NewRegistryFromProviders([]providers.Provider{
|
||||
{Name: "anthropic", Keywords: []string{"sk-ant-"}},
|
||||
})
|
||||
}
|
||||
|
||||
const codesandboxFixtureHTML = `<!doctype html>
|
||||
<html><body>
|
||||
<a href="/s/abc123-def">one</a>
|
||||
<a href="/s/XYZ-789">two</a>
|
||||
<a href="/docs">skip</a>
|
||||
<a href="/s/abc/extra/path">skip deeper</a>
|
||||
</body></html>`
|
||||
|
||||
func newCodeSandboxTestSource(srvURL string) *CodeSandboxSource {
|
||||
return &CodeSandboxSource{
|
||||
BaseURL: srvURL,
|
||||
Registry: codesandboxTestRegistry(),
|
||||
Limiters: recon.NewLimiterRegistry(),
|
||||
Client: NewClient(),
|
||||
}
|
||||
}
|
||||
|
||||
func TestCodeSandbox_Sweep_ExtractsFindings(t *testing.T) {
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||||
if r.URL.Path != "/search" {
|
||||
t.Errorf("unexpected path: %s", r.URL.Path)
|
||||
}
|
||||
if r.URL.Query().Get("query") == "" {
|
||||
t.Errorf("missing query param")
|
||||
}
|
||||
w.Header().Set("Content-Type", "text/html")
|
||||
_, _ = w.Write([]byte(codesandboxFixtureHTML))
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
src := newCodeSandboxTestSource(srv.URL)
|
||||
out := make(chan recon.Finding, 16)
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
|
||||
defer cancel()
|
||||
|
||||
if err := src.Sweep(ctx, "", out); err != nil {
|
||||
t.Fatalf("Sweep err: %v", err)
|
||||
}
|
||||
close(out)
|
||||
|
||||
var findings []recon.Finding
|
||||
for f := range out {
|
||||
findings = append(findings, f)
|
||||
}
|
||||
if len(findings) != 2 {
|
||||
t.Fatalf("expected 2 findings, got %d", len(findings))
|
||||
}
|
||||
for _, f := range findings {
|
||||
if f.SourceType != "recon:codesandbox" {
|
||||
t.Errorf("unexpected SourceType: %s", f.SourceType)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func TestCodeSandbox_RespectsRobots(t *testing.T) {
|
||||
s := &CodeSandboxSource{}
|
||||
if !s.RespectsRobots() {
|
||||
t.Fatal("expected RespectsRobots=true")
|
||||
}
|
||||
}
|
||||
|
||||
func TestCodeSandbox_EnabledAlwaysTrue(t *testing.T) {
|
||||
s := &CodeSandboxSource{}
|
||||
if !s.Enabled(recon.Config{}) {
|
||||
t.Fatal("expected Enabled=true")
|
||||
}
|
||||
}
|
||||
|
||||
func TestCodeSandbox_Sweep_CtxCancelled(t *testing.T) {
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||||
time.Sleep(500 * time.Millisecond)
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
src := newCodeSandboxTestSource(srv.URL)
|
||||
ctx, cancel := context.WithCancel(context.Background())
|
||||
cancel()
|
||||
|
||||
out := make(chan recon.Finding, 4)
|
||||
if err := src.Sweep(ctx, "", out); err == nil {
|
||||
t.Fatal("expected ctx error")
|
||||
}
|
||||
}
|
||||
|
||||
func TestCodeSandbox_NameAndRate(t *testing.T) {
|
||||
s := &CodeSandboxSource{}
|
||||
if s.Name() != "codesandbox" {
|
||||
t.Fatalf("unexpected name: %s", s.Name())
|
||||
}
|
||||
}
|
||||
141
pkg/recon/sources/replit.go
Normal file
141
pkg/recon/sources/replit.go
Normal file
@@ -0,0 +1,141 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"fmt"
|
||||
"io"
|
||||
"net/http"
|
||||
"net/url"
|
||||
"regexp"
|
||||
"time"
|
||||
|
||||
"golang.org/x/net/html"
|
||||
"golang.org/x/time/rate"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
// ReplitSource scrapes replit.com search HTML for references to provider
|
||||
// keywords. Replit has no public search API, so we honor robots.txt and run
|
||||
// at a very conservative 10 req/min rate.
|
||||
//
|
||||
// Emits one Finding per extracted result link, tagged SourceType=recon:replit.
|
||||
type ReplitSource struct {
|
||||
// BaseURL defaults to https://replit.com. Tests override with httptest URL.
|
||||
BaseURL string
|
||||
// Registry drives the keyword query list via BuildQueries.
|
||||
Registry *providers.Registry
|
||||
// Limiters is the shared recon.LimiterRegistry used to coordinate rate.
|
||||
// Callers SweepAll wires this; tests may pass a fresh registry.
|
||||
Limiters *recon.LimiterRegistry
|
||||
// Client is the shared retry HTTP wrapper. If nil, a default is used.
|
||||
Client *Client
|
||||
}
|
||||
|
||||
// replitLinkRE matches /@<user>/<repl-name> result links (no further slashes).
|
||||
var replitLinkRE = regexp.MustCompile(`^/@[^/]+/[^/]+$`)
|
||||
|
||||
// Compile-time assertion that ReplitSource satisfies recon.ReconSource.
|
||||
var _ recon.ReconSource = (*ReplitSource)(nil)
|
||||
|
||||
func (s *ReplitSource) Name() string { return "replit" }
|
||||
func (s *ReplitSource) RateLimit() rate.Limit { return rate.Every(6 * time.Second) }
|
||||
func (s *ReplitSource) Burst() int { return 1 }
|
||||
func (s *ReplitSource) RespectsRobots() bool { return true }
|
||||
|
||||
// Enabled always returns true: Replit scraping requires no credentials.
|
||||
func (s *ReplitSource) Enabled(_ recon.Config) bool { return true }
|
||||
|
||||
// Sweep iterates each provider keyword, performs a Replit search query, and
|
||||
// extracts result anchors from the returned HTML. The engine coordinates
|
||||
// robots.txt at a higher level; per-request rate limiting is delegated to
|
||||
// Limiters when available so individual sweeps play nicely with SweepAll.
|
||||
func (s *ReplitSource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
|
||||
base := s.BaseURL
|
||||
if base == "" {
|
||||
base = "https://replit.com"
|
||||
}
|
||||
client := s.Client
|
||||
if client == nil {
|
||||
client = NewClient()
|
||||
}
|
||||
|
||||
queries := BuildQueries(s.Registry, "replit")
|
||||
if len(queries) == 0 {
|
||||
// No registry → emit nothing, not an error.
|
||||
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/search?q=%s&type=repls", base, url.QueryEscape(q))
|
||||
req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
|
||||
if err != nil {
|
||||
return fmt.Errorf("replit: build req: %w", err)
|
||||
}
|
||||
resp, err := client.Do(ctx, req)
|
||||
if err != nil {
|
||||
return fmt.Errorf("replit: fetch: %w", err)
|
||||
}
|
||||
links, err := extractAnchorHrefs(resp.Body, replitLinkRE)
|
||||
_ = resp.Body.Close()
|
||||
if err != nil {
|
||||
return fmt.Errorf("replit: parse html: %w", err)
|
||||
}
|
||||
|
||||
for _, href := range links {
|
||||
if err := ctx.Err(); err != nil {
|
||||
return err
|
||||
}
|
||||
absURL := base + href
|
||||
out <- recon.Finding{
|
||||
ProviderName: "",
|
||||
Source: absURL,
|
||||
SourceType: "recon:replit",
|
||||
Confidence: "low",
|
||||
DetectedAt: time.Now(),
|
||||
}
|
||||
}
|
||||
}
|
||||
return nil
|
||||
}
|
||||
|
||||
// extractAnchorHrefs walks parsed HTML and returns every <a href> attribute
|
||||
// value whose path matches the given regexp. Duplicates are preserved in order.
|
||||
func extractAnchorHrefs(body io.Reader, re *regexp.Regexp) ([]string, error) {
|
||||
doc, err := html.Parse(body)
|
||||
if err != nil {
|
||||
return nil, err
|
||||
}
|
||||
var out []string
|
||||
seen := make(map[string]struct{})
|
||||
var walk func(*html.Node)
|
||||
walk = func(n *html.Node) {
|
||||
if n.Type == html.ElementNode && n.Data == "a" {
|
||||
for _, a := range n.Attr {
|
||||
if a.Key == "href" && re.MatchString(a.Val) {
|
||||
if _, ok := seen[a.Val]; !ok {
|
||||
seen[a.Val] = struct{}{}
|
||||
out = append(out, a.Val)
|
||||
}
|
||||
break
|
||||
}
|
||||
}
|
||||
}
|
||||
for c := n.FirstChild; c != nil; c = c.NextSibling {
|
||||
walk(c)
|
||||
}
|
||||
}
|
||||
walk(doc)
|
||||
return out, nil
|
||||
}
|
||||
131
pkg/recon/sources/replit_test.go
Normal file
131
pkg/recon/sources/replit_test.go
Normal file
@@ -0,0 +1,131 @@
|
||||
package sources
|
||||
|
||||
import (
|
||||
"context"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"testing"
|
||||
"time"
|
||||
|
||||
"github.com/salvacybersec/keyhunter/pkg/providers"
|
||||
"github.com/salvacybersec/keyhunter/pkg/recon"
|
||||
)
|
||||
|
||||
func replitTestRegistry() *providers.Registry {
|
||||
return providers.NewRegistryFromProviders([]providers.Provider{
|
||||
{Name: "openai", Keywords: []string{"sk-proj-"}},
|
||||
})
|
||||
}
|
||||
|
||||
const replitFixtureHTML = `<!doctype html>
|
||||
<html><body>
|
||||
<a href="/@alice/super-bot">hit one</a>
|
||||
<a href="/@bob/weather-api">hit two</a>
|
||||
<a href="/about">nope</a>
|
||||
<a href="https://external.example.com/nope">external</a>
|
||||
<a href="/@carol/x/nested">too deep nope</a>
|
||||
</body></html>`
|
||||
|
||||
func newReplitTestSource(srvURL string) *ReplitSource {
|
||||
return &ReplitSource{
|
||||
BaseURL: srvURL,
|
||||
Registry: replitTestRegistry(),
|
||||
Limiters: recon.NewLimiterRegistry(),
|
||||
Client: NewClient(),
|
||||
}
|
||||
}
|
||||
|
||||
func TestReplit_Sweep_ExtractsFindings(t *testing.T) {
|
||||
var hits int
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||||
if r.URL.Path != "/search" {
|
||||
t.Errorf("unexpected path: %s", r.URL.Path)
|
||||
}
|
||||
if r.URL.Query().Get("q") == "" {
|
||||
t.Errorf("missing q param")
|
||||
}
|
||||
hits++
|
||||
w.Header().Set("Content-Type", "text/html")
|
||||
_, _ = w.Write([]byte(replitFixtureHTML))
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
src := newReplitTestSource(srv.URL)
|
||||
out := make(chan recon.Finding, 16)
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
|
||||
defer cancel()
|
||||
|
||||
if err := src.Sweep(ctx, "", out); err != nil {
|
||||
t.Fatalf("Sweep err: %v", err)
|
||||
}
|
||||
close(out)
|
||||
|
||||
var findings []recon.Finding
|
||||
for f := range out {
|
||||
findings = append(findings, f)
|
||||
}
|
||||
if len(findings) != 2 {
|
||||
t.Fatalf("expected 2 findings, got %d", len(findings))
|
||||
}
|
||||
wantA := srv.URL + "/@alice/super-bot"
|
||||
wantB := srv.URL + "/@bob/weather-api"
|
||||
got := map[string]bool{}
|
||||
for _, f := range findings {
|
||||
got[f.Source] = true
|
||||
if f.SourceType != "recon:replit" {
|
||||
t.Errorf("unexpected SourceType: %s", f.SourceType)
|
||||
}
|
||||
}
|
||||
if !got[wantA] || !got[wantB] {
|
||||
t.Fatalf("missing expected sources; got=%v", got)
|
||||
}
|
||||
if hits == 0 {
|
||||
t.Fatalf("server was never hit")
|
||||
}
|
||||
}
|
||||
|
||||
func TestReplit_RespectsRobots(t *testing.T) {
|
||||
s := &ReplitSource{}
|
||||
if !s.RespectsRobots() {
|
||||
t.Fatal("expected RespectsRobots=true")
|
||||
}
|
||||
}
|
||||
|
||||
func TestReplit_EnabledAlwaysTrue(t *testing.T) {
|
||||
s := &ReplitSource{}
|
||||
if !s.Enabled(recon.Config{}) {
|
||||
t.Fatal("expected Enabled=true")
|
||||
}
|
||||
}
|
||||
|
||||
func TestReplit_Sweep_CtxCancelled(t *testing.T) {
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||||
time.Sleep(500 * time.Millisecond)
|
||||
_, _ = w.Write([]byte(replitFixtureHTML))
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
src := newReplitTestSource(srv.URL)
|
||||
ctx, cancel := context.WithCancel(context.Background())
|
||||
cancel()
|
||||
|
||||
out := make(chan recon.Finding, 4)
|
||||
if err := src.Sweep(ctx, "", out); err == nil {
|
||||
t.Fatal("expected ctx error")
|
||||
}
|
||||
}
|
||||
|
||||
func TestReplit_NameAndRate(t *testing.T) {
|
||||
s := &ReplitSource{}
|
||||
if s.Name() != "replit" {
|
||||
t.Errorf("unexpected name: %s", s.Name())
|
||||
}
|
||||
if s.Burst() != 1 {
|
||||
t.Errorf("burst: %d", s.Burst())
|
||||
}
|
||||
want := float64(1) / 6
|
||||
got := float64(s.RateLimit())
|
||||
if got < want-0.01 || got > want+0.01 {
|
||||
t.Errorf("rate limit=%v want~%v", got, want)
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user