feat(crawler): redesign crawler scripts and admin workflow

- add generic scriptcrawler backend runner using the crawler.v1 JSONL protocol - support crawler script upload and HTTP(S) URL import from the admin crawler page - simplify the user-facing crawler contract to title, media_url, optional thumbnail_url and optional source_id - convert Spider91 into a normal script crawler and reject new Spider91 storage-drive configs - keep legacy Spider91 storage rows visible only for cleanup/deletion - add crawler protocol docs, example script, admin UI, tests and migration coverage
2026-06-15 08:45:41 +08:00 · 2026-06-09 23:49:28 +08:00
parent 71d4a16db1
commit ec5a01b6aa
25 changed files with 3577 additions and 233 deletions
@@ -12,6 +12,9 @@
    pip install requests beautifulsoup4 lxml PySocks

 使用方法:
+    # 作为 video-site-91 通用爬虫脚本运行（后台会自动这样调用）
+    python spider_91porn.py --job /path/to/job.json
+
    # 全量爬取（默认行为，从 page=1 一直爬到末尾，写到 OUTPUT_FILE）
    python spider_91porn.py

@@ -22,6 +25,7 @@
    python spider_91porn.py --target-new 15 --seen-viewkeys-file /tmp/seen.txt --output /tmp/new.json

 CLI 参数:
+    --job FILE                crawler.v1 job JSON 路径；后台爬虫管理会使用此模式
    --page N                  只爬第 N 页，配合 --output 用于手动调试
    --target-new N            从 page 1 起翻页直到凑够 N 个新视频（不在 seen 列表里的）
    --seen-viewkeys-file FILE 每行一个已知 viewkey 或 mp4 源 ID，命中即跳过；与 --target-new 配合使用
@@ -37,6 +41,8 @@ CLI 参数:
    - OUTPUT_FILE : 输出文件名

 输出格式 (JSON):
+    --job 模式下 stdout 输出 crawler.v1 JSON Lines，日志全部写到 stderr。
+    手动运行模式仍会写传统 JSON 文件：
    {
      "videos": [
        {
@@ -77,8 +83,8 @@ from datetime import datetime
 try:
    from bs4 import BeautifulSoup
 except ImportError:
-    print("错误: 缺少依赖库 beautifulsoup4")
-    print("请运行: pip install beautifulsoup4 lxml")
+    print("错误: 缺少依赖库 beautifulsoup4", file=sys.stderr)
+    print("请运行: pip install beautifulsoup4 lxml", file=sys.stderr)
    sys.exit(1)


@@ -148,9 +154,23 @@ OUTPUT_FILE = "91porn_videos.json"
 MAX_PAGES = None          # 设置为 None 爬取所有页，或设置整数如 5 只爬前5页
 RESUME = True             # 是否跳过输出文件中已存在的 viewkey (断点续爬)
 MAX_EMPTY_PAGES = 2       # 连续空页数达到此值时停止爬取
+CRAWLER_PROTOCOL = "crawler.v1"
 # ===================================================


+def crawler_source_id(raw: str) -> str:
+    """Return a backend-safe source_id, preserving existing numeric 91 IDs."""
+    value = str(raw or "").strip()
+    if not value:
+        return ""
+    safe = re.sub(r"[^A-Za-z0-9_.-]+", "_", value).strip("._-")
+    return safe[:160]
+
+
+def write_jsonl(event: dict):
+    print(json.dumps(event, ensure_ascii=False), flush=True)
+
+
 class Porn91Spider:
    def __init__(
        self,
@@ -163,6 +183,7 @@ class Porn91Spider:
        target_new: int = None,
        seen_viewkeys: list = None,
        stream_output: bool = False,
+        stream_protocol: str = "legacy",
    ):
        """
        构造函数。所有参数都有默认值，等同于使用脚本顶部的全局配置。
@@ -198,6 +219,7 @@ class Porn91Spider:
        # （配合 backend Go 端 bufio.Scanner 实时消费，下载一个就开始下一个）。
        # 开启后所有 log 都走 stderr。
        self.stream_output = bool(stream_output)
+        self.stream_protocol = stream_protocol or "legacy"

        # 添加重试适配器
        try:
@@ -263,7 +285,28 @@ class Porn91Spider:
        if not self.stream_output:
            return
        try:
-            print(json.dumps(video, ensure_ascii=False), flush=True)
+            if self.stream_protocol == "crawler.v1":
+                source_id = crawler_source_id(video.get("source_id") or video.get("viewkey") or "")
+                item = {
+                    "title": video.get("title") or "",
+                    "detail_url": video.get("detail_url") or "",
+                    "author": "91porn",
+                    "tags": ["91porn"],
+                    "media_url": video.get("video_url") or "",
+                    "thumbnail_url": video.get("thumb_url") or "",
+                    "headers": {
+                        "Referer": video.get("detail_url") or BASE_URL,
+                    },
+                }
+                if source_id:
+                    item["source_id"] = source_id
+                event = {
+                    "type": "item",
+                    "item": item,
+                }
+                write_jsonl(event)
+            else:
+                print(json.dumps(video, ensure_ascii=False), flush=True)
        except Exception as e:
            # stdout 异常基本只在管道断开时发生（消费方进程死了）；
            # 写到 stderr 让 backend 看到，然后让 crawl 循环自己 break。
@@ -697,8 +740,9 @@ class Porn91Spider:
        except Exception as e:
            self.log(f"保存文件失败: {e}")
            # 尝试输出到控制台作为备份
-            print("\n--- 备份输出 ---")
-            print(json.dumps(output_data, ensure_ascii=False, indent=2))
+            backup_out = sys.stderr if self.stream_output else sys.stdout
+            print("\n--- 备份输出 ---", file=backup_out, flush=True)
+            print(json.dumps(output_data, ensure_ascii=False, indent=2), file=backup_out, flush=True)

    def _print_summary(self):
        """
@@ -751,6 +795,84 @@ def print_help():
 """)


+def run_job(job_path: str):
+    """Run as a crawler.v1 script plugin.
+
+    The Go host passes a job JSON file and expects stdout JSONL events. Logs go
+    to stderr so stdout stays machine-readable.
+    """
+    with open(job_path, "r", encoding="utf-8") as f:
+        job = json.load(f)
+
+    if job.get("protocol") != CRAWLER_PROTOCOL:
+        raise ValueError(f"unsupported crawler protocol: {job.get('protocol')!r}")
+    if job.get("mode") not in ("", None, "crawl"):
+        raise ValueError(f"unsupported crawler mode: {job.get('mode')!r}")
+
+    try:
+        target_new = int(job.get("target_new") or 15)
+    except (TypeError, ValueError):
+        target_new = 15
+    if target_new <= 0:
+        target_new = 15
+    seen_file = job.get("seen_source_ids_file") or ""
+    output_dir = job.get("output_dir") or os.getcwd()
+    run_id = job.get("run_id") or datetime.utcnow().strftime("%Y%m%dT%H%M%SZ")
+    os.makedirs(output_dir, exist_ok=True)
+    output_file = os.path.join(output_dir, f"spider91-{run_id}.json")
+
+    network = job.get("network") if isinstance(job.get("network"), dict) else {}
+    proxy_url = str(network.get("proxy_url") or "").strip()
+    if proxy_url:
+        os.environ["HTTP_PROXY"] = proxy_url
+        os.environ["HTTPS_PROXY"] = proxy_url
+        os.environ["http_proxy"] = proxy_url
+        os.environ["https_proxy"] = proxy_url
+        os.environ["NO_PROXY"] = ""
+        os.environ["no_proxy"] = ""
+
+    seen_viewkeys = []
+    if seen_file:
+        try:
+            with open(seen_file, "r", encoding="utf-8") as f:
+                for line in f:
+                    line = line.strip()
+                    if line:
+                        seen_viewkeys.append(line)
+        except FileNotFoundError:
+            print(f"警告: seen_source_ids_file 不存在: {seen_file}", file=sys.stderr, flush=True)
+        except Exception as e:
+            print(f"警告: 读取 seen_source_ids_file 失败: {e}", file=sys.stderr, flush=True)
+
+    prefer_ipv4_for_plain_socks5_proxy()
+    spider = Porn91Spider(
+        output_file=output_file,
+        start_page=1,
+        max_pages=None,
+        resume=False,
+        quiet=True,
+        target_new=target_new,
+        seen_viewkeys=seen_viewkeys,
+        stream_output=True,
+        stream_protocol="crawler.v1",
+    )
+    try:
+        spider.crawl()
+        done = {
+            "type": "done",
+            "stats": {
+                "emitted": spider.processed_videos,
+                "failed": spider.failed_videos,
+                "skipped": spider.skipped_videos,
+            },
+        }
+        write_jsonl(done)
+    except KeyboardInterrupt:
+        spider.log("\n用户中断，正在保存已爬取的数据...")
+        spider._save_results()
+        raise
+
+
 def main():
    if len(sys.argv) > 1 and sys.argv[1] in ('-h', '--help', 'help'):
        print_help()
@@ -778,8 +900,14 @@ def main():
    parser.add_argument("--stream-output", action="store_true",
                        help="流式模式：每解析一条视频直链就立即把它作为一行 JSON 写到 stdout 并 flush；"
                             "日志改走 stderr。配合 backend 边读边下载使用。")
+    parser.add_argument("--job", type=str, default=None,
+                        help="crawler.v1 job JSON 路径；作为通用脚本爬虫运行。")

    args, _ = parser.parse_known_args()
+    if args.job:
+        run_job(args.job)
+        return
+
    cli_out = sys.stderr if args.stream_output else sys.stdout
    prefer_ipv4_for_plain_socks5_proxy()

@@ -23,6 +23,7 @@ internal/
    onedrive/               OneDrive（OpenList 在线续期 + Microsoft Graph 文件接口）
    googledrive/            Google Drive（OpenList 在线续期 + Google Drive API；播放走后端代理）
    localstorage/           本地目录扫描（服务器已有视频目录）
+    scriptcrawler/          通用脚本爬虫输出的本地媒体适配层
  scanner/                  扫目录 → 落库
  preview/                  ffmpeg 抽封面和生成多段预览视频
  proxy/                    /p/stream/*、/p/preview/* 代理
@@ -79,6 +80,12 @@ npm run preview     前端 9191，无热更新
 go run ./cmd/server 后端 9192
 ```

+## 爬虫脚本
+
+爬虫现在是独立后台栏目 `/admin/crawlers`，不再作为“网盘/存储类型”配置。脚本负责发现视频，后端负责去重、下载、入库、封面、预览视频和视频指纹。
+
+脚本协议见 [docs/crawler-protocol.md](../docs/crawler-protocol.md)。后台支持上传 `.py` 文件或通过 HTTP(S) 脚本链接导入，导入后的脚本会保存到数据目录旁的 `crawler-scripts/`。内置 91 爬虫也支持同一套 `crawler.v1` job 协议；后台“内置 91”会自动使用仓库里的 `91VideoSpider/spider_91porn.py`。
+
 ## 添加一个盘

 推荐在前端管理后台 `/admin/drives` 新增网盘。保存后会立即挂载并触发扫描；视频结果可在 `/admin/videos` 按网盘查看，每页 100 条，页面会同时显示各网盘预览视频已生成、待生成、失败数量。
@@ -33,6 +33,7 @@ import (
 	"github.com/video-site/backend/internal/drives/p123"
 	"github.com/video-site/backend/internal/drives/pikpak"
 	"github.com/video-site/backend/internal/drives/quark"
+	"github.com/video-site/backend/internal/drives/scriptcrawler"
 	"github.com/video-site/backend/internal/drives/spider91"
 	"github.com/video-site/backend/internal/drives/wopan"
 	"github.com/video-site/backend/internal/fingerprint"
@@ -45,6 +46,7 @@ import (
 )

 const fingerprintReconcileInterval = time.Minute
+const legacySpider91DriveUnsupported = "91Spider 已不再支持作为网盘配置，请在爬虫管理页面添加爬虫脚本"

 func main() {
 	cfgPath := "./config.yaml"
@@ -76,7 +78,7 @@ func main() {
 		workers:            make(map[string]*preview.Worker),
 		thumbWorkers:       make(map[string]*preview.ThumbWorker),
 		fingerprintWorkers: make(map[string]*fingerprint.Worker),
-		spider91Crawlers:   make(map[string]*spider91.Crawler),
+		scriptCrawlers:     make(map[string]*scriptcrawler.Crawler),
 	}
 	app.proxy = proxy.New(app.registry)
 	app.spider91Migrator = spider91migrate.New(spider91migrate.Config{
@@ -171,13 +173,23 @@ func main() {
 			app.detachDrive(driveID)
 		},
 		OnScanRequested: func(driveID string) bool {
-			// spider91 的"重扫"等同于手动触发一次爬取；其它 drive 走标准 scan
-			app.mu.Lock()
-			_, isSpider91 := app.spider91Crawlers[driveID]
-			app.mu.Unlock()
+			// 爬虫类 drive 的"重扫"等同于手动触发一次爬取；其它 drive 走标准 scan
+			isSpider91 := false
+			isScriptCrawler := false
+			if d, err := app.cat.GetDrive(ctx, driveID); err == nil && d != nil {
+				if d.Kind == spider91.Kind {
+					log.Printf("[spider91] drive=%s is a deprecated storage crawler, ignore scan request", driveID)
+					return false
+				}
+				isSpider91 = scriptCrawlerSourceKindForDrive(d) == spider91.Kind
+				isScriptCrawler = d.Kind == scriptcrawler.Kind
+			}
 			if isSpider91 {
 				return app.scheduleSpider91Crawl(ctx, driveID)
 			}
+			if isScriptCrawler {
+				return app.scheduleScriptCrawlerCrawl(ctx, driveID)
+			}
 			return app.scheduleScan(ctx, driveID)
 		},
 		OnStopDriveTasks: func(driveID string) bool {
@@ -227,6 +239,9 @@ func main() {
 		SetSpider91UploadDriveID: func(id string) error {
 			return app.SetSpider91UploadDriveID(ctx, id)
 		},
+		DefaultSpider91ScriptPath: func() string {
+			return app.defaultSpider91ScriptPath()
+		},
 		OnRunNightlyJob: func() bool {
 			if app.nightlyRunner != nil {
 				return app.nightlyRunner.TriggerNow()
@@ -304,8 +319,9 @@ type App struct {
 	thumbWorkers       map[string]*preview.ThumbWorker
 	fingerprintWorkers map[string]*fingerprint.Worker
 	cancels            map[string]context.CancelFunc
-	// spider91Crawlers 按 driveID 索引，每个 spider91 drive 独立一个 Crawler
-	spider91Crawlers map[string]*spider91.Crawler
+	// scriptCrawlers 按 driveID 索引，每个脚本爬虫 drive 独立一个 Crawler。
+	// 内置 Spider91 也走这里，只是 SourceKind=spider91，以兼容历史 video id。
+	scriptCrawlers map[string]*scriptcrawler.Crawler

 	// driveAttachMu 串行化云盘挂载/重挂载。挂载会访问上游服务，可能较慢；
 	// 串行化可以避免启动后台挂载和手动扫盘按需挂载同一个 drive 时重复创建 worker。
@@ -737,11 +753,16 @@ func (a *App) attachDriveUnlocked(ctx context.Context, d *catalog.Drive) error {
 			ID:       d.ID,
 			RootPath: d.Credentials["path"],
 		})
-	case spider91.Kind:
-		drv = spider91.New(spider91.Config{
+	case scriptcrawler.Kind:
+		drv = scriptcrawler.New(scriptcrawler.Config{
 			ID:      d.ID,
-			RootDir: a.spider91DriveDir(d.ID),
+			RootDir: a.scriptCrawlerDriveDirForDrive(d),
 		})
+	case spider91.Kind:
+		d.Status = "error"
+		d.LastError = legacySpider91DriveUnsupported
+		_ = a.cat.UpsertDrive(ctx, d)
+		return errors.New(legacySpider91DriveUnsupported)
 	default:
 		return fmt.Errorf("unknown drive kind: %s", d.Kind)
 	}
@@ -761,9 +782,8 @@ func (a *App) attachDriveUnlocked(ctx context.Context, d *catalog.Drive) error {

 	a.startDriveGenerationWorkers(ctx, d.ID, drv, true)

-	// spider91 driver 还需要一个 crawler，挂在专用 map 里供 crawlerLoop 调用
-	if sd, ok := drv.(*spider91.Driver); ok {
-		a.attachSpider91Crawler(d, sd)
+	if sd, ok := drv.(*scriptcrawler.Driver); ok {
+		a.attachScriptCrawler(d, sd)
 	}

 	return nil
@@ -836,6 +856,26 @@ func (a *App) spider91DriveDir(driveID string) string {
 	return filepath.Join(a.spider91RootDir(), driveID)
 }

+// scriptCrawlerRootDir 是所有通用脚本爬虫 drive 共享的根目录。
+func (a *App) scriptCrawlerRootDir() string {
+	return filepath.Join(filepath.Dir(a.cfg.Storage.LocalPreviewDir), "scriptcrawlers")
+}
+
+// scriptCrawlerDriveDir 是单个 scriptcrawler drive 的存储目录：<root>/<driveID>。
+func (a *App) scriptCrawlerDriveDir(driveID string) string {
+	return filepath.Join(a.scriptCrawlerRootDir(), driveID)
+}
+
+func (a *App) scriptCrawlerDriveDirForDrive(d *catalog.Drive) string {
+	if d != nil && scriptCrawlerSourceKindForDrive(d) == spider91.Kind {
+		return a.spider91DriveDir(d.ID)
+	}
+	if d == nil {
+		return a.scriptCrawlerDriveDir("")
+	}
+	return a.scriptCrawlerDriveDir(d.ID)
+}
+
 // commonThumbsDir 是所有 drive 共享的封面目录，/p/thumb/{videoID} 路由命中这里。
 func (a *App) commonThumbsDir() string {
 	return filepath.Join(a.cfg.Storage.LocalPreviewDir, "thumbs")
@@ -865,77 +905,72 @@ func (a *App) defaultSpider91ScriptPath() string {
 	return ""
 }

-// attachSpider91Crawler 创建该 drive 对应的 Crawler 并注册到 a.spider91Crawlers。
-func (a *App) attachSpider91Crawler(d *catalog.Drive, drv *spider91.Driver) {
+// attachScriptCrawler 创建通用脚本爬虫 runner，并注册到 a.scriptCrawlers。
+func (a *App) attachScriptCrawler(d *catalog.Drive, drv *scriptcrawler.Driver) {
 	pythonPath := strings.TrimSpace(d.Credentials["python_path"])
 	if pythonPath == "" {
 		pythonPath = "python3"
 	}
 	scriptPath := strings.TrimSpace(d.Credentials["script_path"])
-	if scriptPath == "" {
+	sourceKind := scriptCrawlerSourceKindForDrive(d)
+	if scriptPath == "" && sourceKind == spider91.Kind {
 		scriptPath = a.defaultSpider91ScriptPath()
 	}
-	// 91porn CDN 在海外；空缺时回退到 HTTPS_PROXY / HTTP_PROXY 环境变量。
 	proxyURL := strings.TrimSpace(d.Credentials["proxy"])
+	configJSON := strings.TrimSpace(d.Credentials["config_json"])
+	workDir := ""
+	if scriptPath != "" {
+		workDir = filepath.Dir(scriptPath)
+	}

 	driveID := d.ID
-	var progressMu sync.Mutex
-	checkedVideos := 0
-	expectedNewVideos := 0
-	updateProgress := func(scanned, added int) {
-		a.updateDriveScanProgress(driveID, scanned, added)
-	}
-	c := spider91.NewCrawler(spider91.CrawlerConfig{
+	c := scriptcrawler.NewCrawler(scriptcrawler.CrawlerConfig{
 		Driver:         drv,
 		Catalog:        a.cat,
+		SourceKind:     sourceKind,
 		PythonPath:     pythonPath,
 		ScriptPath:     scriptPath,
-		WorkDir:        filepath.Dir(scriptPath),
+		WorkDir:        workDir,
 		CommonThumbDir: a.commonThumbsDir(),
 		ProxyURL:       proxyURL,
-		OnProgress: func(progress spider91.CrawlProgress) {
-			progressMu.Lock()
-			if progress.TotalEntries == 0 && progress.NewVideos == 0 && progress.Skipped == 0 && progress.Failed == 0 {
-				checkedVideos = 0
-				expectedNewVideos = 0
-			} else if progress.TotalEntries > expectedNewVideos {
-				expectedNewVideos = progress.TotalEntries
+		ConfigJSON:     configJSON,
+		OnProgress: func(progress scriptcrawler.CrawlProgress) {
+			scanned := progress.Checked
+			if scanned < progress.TotalEntries {
+				scanned = progress.TotalEntries
 			}
-			scanned := checkedVideos
-			added := expectedNewVideos
-			progressMu.Unlock()
-			updateProgress(scanned, added)
+			added := progress.Emitted
+			if added < progress.NewVideos {
+				added = progress.NewVideos
+			}
+			a.updateDriveScanProgress(driveID, scanned, added)
 		},
-		OnCheckedVideo: func() {
-			progressMu.Lock()
-			checkedVideos++
-			scanned := checkedVideos
-			added := expectedNewVideos
-			progressMu.Unlock()
-			updateProgress(scanned, added)
-		},
-		OnExtractedVideo: func() {
-			progressMu.Lock()
-			expectedNewVideos++
-			scanned := checkedVideos
-			added := expectedNewVideos
-			progressMu.Unlock()
-			updateProgress(scanned, added)
-		},
-		// 新流程：预览视频不在每条视频入库时立即入队，而是 RunOnce 全部下完后由
-		// runSpider91Crawl 统一调 enqueueDriveGeneration 一次性入队。这样：
-		//   - 下载阶段不和 ffmpeg 抢 CPU/IO
-		//   - "等待预览视频队列 idle" 在 nightly Phase 2 的语义上更直观
-		// 不再传 OnNewVideo（crawler 内部的回调字段保留，仅为单测计数器之用）。
 	})

 	a.mu.Lock()
-	a.spider91Crawlers[driveID] = c
+	a.scriptCrawlers[driveID] = c
 	a.mu.Unlock()

-	// 确保 "91porn" 系统标签存在，并按 spider91 来源前缀给历史视频补打。
-	// 不能只靠文本匹配：老版本入库的视频可能没有 author/tags 字段，但 id 前缀
-	// "spider91-<driveID>-" 会一直保留，即使后续迁移到 PikPak/115 也不变。
+	if sourceKind == spider91.Kind {
+		a.ensureSpider91SourceTag(driveID)
+	}
+}
+
+func scriptCrawlerSourceKindForDrive(d *catalog.Drive) string {
+	if d == nil {
+		return scriptcrawler.Kind
+	}
+	if d.Kind == scriptcrawler.Kind && strings.EqualFold(strings.TrimSpace(d.Credentials["builtin"]), spider91.Kind) {
+		return spider91.Kind
+	}
+	return scriptcrawler.Kind
+}
+
+func isSpider91SourceDrive(d *catalog.Drive) bool {
+	return d != nil && (strings.EqualFold(d.Kind, spider91.Kind) || scriptCrawlerSourceKindForDrive(d) == spider91.Kind)
+}
+
+func (a *App) ensureSpider91SourceTag(driveID string) {
 	bgCtx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
 	go func() {
 		defer cancel()
@@ -1455,7 +1490,7 @@ func (a *App) detachDrive(id string) {
 	delete(a.workers, id)
 	delete(a.thumbWorkers, id)
 	delete(a.fingerprintWorkers, id)
-	delete(a.spider91Crawlers, id)
+	delete(a.scriptCrawlers, id)
 	a.mu.Unlock()
 }

@@ -1611,7 +1646,7 @@ func (a *App) runScanWithTaskContext(ctx context.Context, driveID string) {
 	// spider91 / localupload 走自己的生命周期管理，不应该参与扫描清理；
 	// stats.Errors > 0 时（云盘 API 中途抖动）保守起见跳过这一轮，避免把
 	// "暂时列不出来"误认成"被用户删了"。
-	if drv.Kind() != spider91.Kind && drv.ID() != localupload.DriveID {
+	if drv.Kind() != spider91.Kind && drv.Kind() != scriptcrawler.Kind && drv.ID() != localupload.DriveID {
 		if stats.Errors > 0 {
 			log.Printf("[cleanup] skip stale cleanup for drive=%s kind=%s: scan had %d directory errors", driveID, drv.Kind(), stats.Errors)
 		} else {
@@ -1736,7 +1771,7 @@ func (a *App) spider91OriginFromVideo(ctx context.Context, v *catalog.Video) (st
 	if a == nil || v == nil {
 		return "", ""
 	}
-	if d, err := a.cat.GetDrive(ctx, v.DriveID); err == nil && d != nil && d.Kind == spider91.Kind {
+	if d, err := a.cat.GetDrive(ctx, v.DriveID); err == nil && d != nil && isSpider91SourceDrive(d) {
 		prefix := "spider91-" + d.ID + "-"
 		if strings.HasPrefix(v.ID, prefix) {
 			return d.ID, strings.TrimPrefix(v.ID, prefix)
@@ -1749,7 +1784,7 @@ func (a *App) spider91OriginFromVideo(ctx context.Context, v *catalog.Video) (st
 	bestDriveID := ""
 	bestSourceID := ""
 	for _, d := range drives {
-		if d == nil || d.Kind != spider91.Kind {
+		if d == nil || !isSpider91SourceDrive(d) {
 			continue
 		}
 		prefix := "spider91-" + d.ID + "-"
@@ -1839,7 +1874,7 @@ func (a *App) cleanupDriveVideosForDelete(ctx context.Context, driveID string) (
 		}
 	}

-	if strings.EqualFold(d.Kind, spider91.Kind) {
+	if isSpider91SourceDrive(d) {
 		if err := a.removeSpider91DriveDir(driveID); err != nil {
 			return 0, err
 		}
@@ -1924,7 +1959,7 @@ func (a *App) videosForDriveDelete(ctx context.Context, d *catalog.Drive) ([]*ca
 		byID[v.ID] = v
 	}

-	if strings.EqualFold(d.Kind, spider91.Kind) {
+	if isSpider91SourceDrive(d) {
 		prefix := "spider91-" + d.ID + "-"
 		originItems, err := a.cat.ListVideosByIDPrefix(ctx, prefix)
 		if err != nil {
@@ -2380,7 +2415,7 @@ func (a *App) regenFailedFingerprints(ctx context.Context, driveID string) {
 }

 // listScanTargetIDs 返回 nightly Phase 1 应扫描的所有 drive ID
-// （非 spider91、非 localupload）。它直接读 catalog，而不是 registry，这样
+// （非爬虫、非 localupload）。它直接读 catalog，而不是 registry，这样
 // 进程刚启动、云盘还在后台挂载时，nightly 也不会漏掉配置过的 drive。
 func (a *App) listScanTargetIDs(ctx context.Context) []string {
 	all, err := a.cat.ListDrives(ctx)
@@ -2390,7 +2425,7 @@ func (a *App) listScanTargetIDs(ctx context.Context) []string {
 	}
 	out := make([]string, 0, len(all))
 	for _, d := range all {
-		if d == nil || d.ID == localupload.DriveID || d.Kind == spider91.Kind {
+		if d == nil || d.ID == localupload.DriveID || d.Kind == spider91.Kind || d.Kind == scriptcrawler.Kind {
 			continue
 		}
 		out = append(out, d.ID)
@@ -2398,7 +2433,7 @@ func (a *App) listScanTargetIDs(ctx context.Context) []string {
 	return out
 }

-// listSpider91DriveIDs 返回 nightly Phase 2 应触发爬取的 spider91 drive ID 列表。
+// listSpider91DriveIDs 返回 nightly Phase 2 应触发爬取的爬虫 drive ID 列表。
 func (a *App) listSpider91DriveIDs(ctx context.Context) []string {
 	all, err := a.cat.ListDrives(ctx)
 	if err != nil {
@@ -2407,7 +2442,7 @@ func (a *App) listSpider91DriveIDs(ctx context.Context) []string {
 	}
 	out := make([]string, 0, len(all))
 	for _, d := range all {
-		if d != nil && d.Kind == spider91.Kind {
+		if d != nil && d.Kind == scriptcrawler.Kind {
 			out = append(out, d.ID)
 		}
 	}
@@ -2449,8 +2484,8 @@ func shouldScanDrive(d drives.Drive) bool {
 	if d == nil || d.ID() == localupload.DriveID {
 		return false
 	}
-	// spider91 由专用的 crawlerLoop 触发，不参与 scanLoop
-	if d.Kind() == spider91.Kind {
+	// 爬虫类 drive 由专用 crawl 阶段触发，不参与普通 scan
+	if d.Kind() == spider91.Kind || d.Kind() == scriptcrawler.Kind {
 		return false
 	}
 	return true
@@ -2481,65 +2516,96 @@ func (a *App) scheduleSpider91Crawl(ctx context.Context, driveID string) bool {
 	return true
 }

+func (a *App) scheduleScriptCrawlerCrawl(ctx context.Context, driveID string) bool {
+	if a.driveHasActiveWork(driveID) {
+		log.Printf("[scriptcrawler] drive=%s has active work, skip duplicate crawl request", driveID)
+		return false
+	}
+	if !a.beginDriveScanOrCrawl(driveID) {
+		log.Printf("[scriptcrawler] drive=%s already queued or running, skip duplicate crawl request", driveID)
+		return false
+	}
+	taskCtx, done := a.registerDriveTaskContext(ctx, driveID)
+
+	go func() {
+		defer func() {
+			a.endDriveScanOrCrawl(driveID)
+			done()
+		}()
+		a.runScriptCrawlerCrawlWithTaskContext(taskCtx, driveID)
+	}()
+	return true
+}
+
 // runSpider91Crawl 运行一次完整爬取流程并把 last_crawl_at 写回 drive.credentials。
 //
 // 即使爬取失败也会更新 last_crawl_at，避免一直在错误循环里反复触发；下一次 nightly
 // 流水线重跑时仍会重试。该方法是阻塞的，被 nightly Phase 2 串行调用，以及被
 // admin "立即抓取" 单 drive 异步调用。
 func (a *App) runSpider91Crawl(ctx context.Context, driveID string) {
+	a.runScriptCrawlerCrawl(ctx, driveID)
+}
+
+func (a *App) runScriptCrawlerCrawl(ctx context.Context, driveID string) {
 	if !a.beginDriveScanOrCrawl(driveID) {
-		log.Printf("[spider91] drive=%s already queued or running, skip direct crawl", driveID)
+		log.Printf("[scriptcrawler] drive=%s already queued or running, skip direct crawl", driveID)
 		return
 	}
 	defer a.endDriveScanOrCrawl(driveID)
 	taskCtx, done := a.registerDriveTaskContext(ctx, driveID)
 	defer done()
-	a.runSpider91CrawlWithTaskContext(taskCtx, driveID)
+	a.runScriptCrawlerCrawlWithTaskContext(taskCtx, driveID)
 }

 func (a *App) runSpider91CrawlWithTaskContext(ctx context.Context, driveID string) bool {
+	return a.runScriptCrawlerCrawlWithTaskContext(ctx, driveID)
+}
+
+func (a *App) runScriptCrawlerCrawlWithTaskContext(ctx context.Context, driveID string) bool {
 	if err := ctx.Err(); err != nil {
-		log.Printf("[spider91] drive=%s crawl canceled before start: %v", driveID, err)
+		log.Printf("[scriptcrawler] drive=%s crawl canceled before start: %v", driveID, err)
 		return false
 	}
 	a.mu.Lock()
-	c := a.spider91Crawlers[driveID]
+	c := a.scriptCrawlers[driveID]
 	a.mu.Unlock()
 	if c == nil {
 		if err := a.ensureDriveAttached(ctx, driveID); err != nil {
-			log.Printf("[spider91] drive=%s attach failed: %v", driveID, err)
+			log.Printf("[scriptcrawler] drive=%s attach failed: %v", driveID, err)
 			return false
 		}
 		a.mu.Lock()
-		c = a.spider91Crawlers[driveID]
+		c = a.scriptCrawlers[driveID]
 		a.mu.Unlock()
 		if c == nil {
-			log.Printf("[spider91] drive=%s crawler not attached", driveID)
+			log.Printf("[scriptcrawler] drive=%s crawler not attached", driveID)
 			return false
 		}
 	}

 	d, err := a.cat.GetDrive(ctx, driveID)
 	if err != nil || d == nil {
-		log.Printf("[spider91] drive=%s lookup failed: %v", driveID, err)
+		log.Printf("[scriptcrawler] drive=%s lookup failed: %v", driveID, err)
 		return false
 	}
-	targetNew := spider91IntCred(d, "target_new", spider91.DefaultTargetNew)
+	defaultTargetNew := scriptcrawler.DefaultTargetNew
+	if scriptCrawlerSourceKindForDrive(d) == spider91.Kind {
+		defaultTargetNew = spider91.DefaultTargetNew
+	}
+	targetNew := spider91IntCred(d, "target_new", defaultTargetNew)
 	if targetNew <= 0 {
-		targetNew = spider91.DefaultTargetNew
+		targetNew = defaultTargetNew
 	}

-	log.Printf("[spider91] drive=%s start crawl target_new=%d", driveID, targetNew)
+	log.Printf("[scriptcrawler] drive=%s start crawl target_new=%d", driveID, targetNew)
 	res, runErr := c.RunOnce(ctx, targetNew)
 	if runErr != nil {
-		log.Printf("[spider91] drive=%s crawl failed: %v", driveID, runErr)
+		log.Printf("[scriptcrawler] drive=%s crawl failed: %v", driveID, runErr)
 	} else if res != nil {
-		log.Printf("[spider91] drive=%s crawl done target=%d total=%d new=%d skipped=%d failed=%d seen_snapshot=%d",
+		log.Printf("[scriptcrawler] drive=%s crawl done target=%d total=%d new=%d skipped=%d failed=%d seen_snapshot=%d",
 			driveID, res.TargetNew, res.TotalEntries, res.NewVideos, res.Skipped, res.Failed, res.SeenSnapshot)
 	}

-	// 标记最后一次爬取时间。这字段已不再用于调度判定（nightly 流水线统一调度），
-	// 留着仅作为 admin UI 显示"上次抓取 N 小时前"用。
 	if d.Credentials == nil {
 		d.Credentials = make(map[string]string)
 	}
@@ -2552,18 +2618,13 @@ func (a *App) runSpider91CrawlWithTaskContext(ctx context.Context, driveID strin
 		d.LastError = ""
 	}
 	if err := a.cat.UpsertDrive(ctx, d); err != nil {
-		log.Printf("[spider91] drive=%s update last_crawl_at: %v", driveID, err)
+		log.Printf("[scriptcrawler] drive=%s update last_crawl_at: %v", driveID, err)
 	}
 	if err := ctx.Err(); err != nil {
-		log.Printf("[spider91] drive=%s crawl canceled after run: %v", driveID, err)
+		log.Printf("[scriptcrawler] drive=%s crawl canceled after run: %v", driveID, err)
 		return false
 	}

-	// 爬取全部完成后，统一把所有还 pending 的预览视频入队。
-	// 这是新流水线设计：crawler 自身不再每条入库就立即触发预览视频生成，
-	// 让"下载阶段"和"预览视频阶段"在时间上分清楚（也跟 nightly Phase 2
-	// 的"等预览视频队列 idle"语义对齐）。enqueueDriveGeneration 内部会读
-	// 该 drive 当前的 teaser_enabled，关闭时是 noop。
 	a.mu.Lock()
 	worker := a.workers[driveID]
 	thumbWorker := a.thumbWorkers[driveID]
@@ -6,6 +6,7 @@ import (
 	"io"
 	"os"
 	"path/filepath"
+	"strings"
 	"sync"
 	"testing"
 	"time"
@@ -13,6 +14,7 @@ import (
 	"github.com/video-site/backend/internal/catalog"
 	"github.com/video-site/backend/internal/config"
 	"github.com/video-site/backend/internal/drives"
+	"github.com/video-site/backend/internal/drives/scriptcrawler"
 	"github.com/video-site/backend/internal/drives/spider91"
 	"github.com/video-site/backend/internal/fingerprint"
 	"github.com/video-site/backend/internal/preview"
@@ -606,7 +608,8 @@ func TestNightlyTargetsComeFromCatalogBeforeDriveAttach(t *testing.T) {
 	for _, d := range []*catalog.Drive{
 		{ID: "115", Kind: "p115", Name: "115", RootID: "0", TeaserEnabled: true},
 		{ID: "pikpak", Kind: "pikpak", Name: "PikPak", RootID: "0", TeaserEnabled: true},
-		{ID: "91-spider", Kind: "spider91", Name: "91 Spider", RootID: "0", TeaserEnabled: true},
+		{ID: "91-legacy", Kind: "spider91", Name: "91 Legacy", RootID: "0", TeaserEnabled: true},
+		{ID: "91-crawler", Kind: scriptcrawler.Kind, Name: "91 Spider", RootID: "/", TeaserEnabled: true},
 	} {
 		if err := cat.UpsertDrive(ctx, d); err != nil {
 			t.Fatalf("seed drive %s: %v", d.ID, err)
@@ -619,8 +622,47 @@ func TestNightlyTargetsComeFromCatalogBeforeDriveAttach(t *testing.T) {
 		t.Fatalf("scan target ids = %#v, want 115 and pikpak from catalog", scanIDs)
 	}
 	spiderIDs := app.listSpider91DriveIDs(ctx)
-	if len(spiderIDs) != 1 || spiderIDs[0] != "91-spider" {
-		t.Fatalf("spider91 ids = %#v, want catalog spider drive", spiderIDs)
+	if len(spiderIDs) != 1 || spiderIDs[0] != "91-crawler" {
+		t.Fatalf("spider91 ids = %#v, want crawler-page script drive", spiderIDs)
+	}
+}
+
+func TestAttachDriveRejectsLegacySpider91Storage(t *testing.T) {
+	ctx := context.Background()
+	cat, err := catalog.Open(t.TempDir() + "/catalog.db")
+	if err != nil {
+		t.Fatalf("open catalog: %v", err)
+	}
+	t.Cleanup(func() {
+		if err := cat.Close(); err != nil {
+			t.Fatalf("close catalog: %v", err)
+		}
+	})
+	d := &catalog.Drive{
+		ID:            "91-legacy",
+		Kind:          spider91.Kind,
+		Name:          "91 Legacy",
+		RootID:        "/",
+		TeaserEnabled: true,
+	}
+	if err := cat.UpsertDrive(ctx, d); err != nil {
+		t.Fatalf("seed drive: %v", err)
+	}
+
+	app := &App{cat: cat, registry: proxy.NewRegistry()}
+	err = app.attachDrive(ctx, d)
+	if err == nil || !strings.Contains(err.Error(), "爬虫管理") {
+		t.Fatalf("attach err = %v, want crawler management guidance", err)
+	}
+	if _, ok := app.registry.Get(d.ID); ok {
+		t.Fatal("legacy spider91 drive should not be registered")
+	}
+	got, err := cat.GetDrive(ctx, d.ID)
+	if err != nil {
+		t.Fatalf("get drive: %v", err)
+	}
+	if got.Status != "error" || !strings.Contains(got.LastError, "爬虫管理") {
+		t.Fatalf("status/error = %q/%q, want deprecated error", got.Status, got.LastError)
 	}
 }

@@ -1033,7 +1075,6 @@ func TestCleanupDriveVideosForDeleteRemovesRowsAndGeneratedAssetsOnly(t *testing
 		workers:            make(map[string]*preview.Worker),
 		thumbWorkers:       make(map[string]*preview.ThumbWorker),
 		fingerprintWorkers: make(map[string]*fingerprint.Worker),
-		spider91Crawlers:   make(map[string]*spider91.Crawler),
 	}
 	removed, err := app.cleanupDriveVideosForDelete(ctx, "local-main")
 	if err != nil {
@@ -1313,7 +1354,6 @@ func TestCleanupDriveVideosForDeleteSpider91RemovesCrawledDirAndOriginRecords(t
 		workers:            make(map[string]*preview.Worker),
 		thumbWorkers:       make(map[string]*preview.ThumbWorker),
 		fingerprintWorkers: make(map[string]*fingerprint.Worker),
-		spider91Crawlers:   make(map[string]*spider91.Crawler),
 	}
 	removed, err := app.cleanupDriveVideosForDelete(ctx, driveID)
 	if err != nil {
@@ -10,6 +10,8 @@ import (
 	"net/http"
 	"net/url"
 	"os"
+	"path"
+	"path/filepath"
 	"strconv"
 	"strings"
 	"time"
@@ -19,6 +21,7 @@ import (
 	"github.com/video-site/backend/internal/auth"
 	"github.com/video-site/backend/internal/catalog"
 	"github.com/video-site/backend/internal/drives/p123"
+	"github.com/video-site/backend/internal/drives/scriptcrawler"
 )

 type AdminServer struct {
@@ -65,6 +68,9 @@ type AdminServer struct {
 	// Spider91 → 115/123/PikPak/OneDrive 上传目标 drive ID 读写
 	GetSpider91UploadDriveID func() string
 	SetSpider91UploadDriveID func(driveID string) error
+	// DefaultSpider91ScriptPath returns the built-in Spider91 crawler script
+	// path for the independent crawler management UI.
+	DefaultSpider91ScriptPath func() string
 	// OnRunNightlyJob 触发一次完整的凌晨流水线（Phase1 扫盘 + Phase2 91 爬虫 +
 	// Phase3 迁移）。立即返回 —— 实际任务在后台跑，admin 在日志或下次状态查询里
 	// 看进度。若流水线正在跑或已排队，Runner 会拒绝重复触发。
@@ -116,6 +122,8 @@ type NightlyJobStatus struct {
 	LastFinishedAt string `json:"lastFinishedAt,omitempty"`
 }

+const maxCrawlerScriptBytes = 2 * 1024 * 1024
+
 type DeleteVideoResult struct {
 	OK            bool `json:"ok"`
 	DeletedSource bool `json:"deletedSource"`
@@ -150,6 +158,15 @@ func (a *AdminServer) Register(r chi.Router) {
 			r.Post("/drives/{id}/thumbnails/failed/regenerate", a.handleRegenFailedThumbnails)
 			r.Post("/drives/{id}/fingerprints/failed/regenerate", a.handleRegenFailedFingerprints)

+			// 爬虫
+			r.Get("/crawlers", a.handleListCrawlers)
+			r.Post("/crawlers", a.handleUpsertCrawler)
+			r.Post("/crawlers/import-file", a.handleImportCrawlerScriptFile)
+			r.Post("/crawlers/import-url", a.handleImportCrawlerScriptURL)
+			r.Delete("/crawlers/{id}", a.handleDeleteCrawler)
+			r.Post("/crawlers/{id}/run", a.handleRunCrawler)
+			r.Post("/crawlers/{id}/tasks/stop", a.handleStopCrawlerTasks)
+
 			// 视频
 			r.Get("/videos", a.handleAdminListVideos)
 			r.Put("/videos/{id}", a.handleUpdateVideo)
@@ -424,6 +441,11 @@ func (a *AdminServer) handleListDrives(w http.ResponseWriter, r *http.Request) {
 		// LastCrawlAt 是 spider91 上次成功爬取的 unix 秒（来自 credentials.last_crawl_at）。
 		// 其它 kind 留 0；前端用它显示"上次抓取: N 小时前"。
 		Spider91Proxy                 string           `json:"spider91Proxy,omitempty"`
+		ScriptCrawlerPythonPath       string           `json:"scriptCrawlerPythonPath,omitempty"`
+		ScriptCrawlerScriptPath       string           `json:"scriptCrawlerScriptPath,omitempty"`
+		ScriptCrawlerProxy            string           `json:"scriptCrawlerProxy,omitempty"`
+		ScriptCrawlerTargetNew        string           `json:"scriptCrawlerTargetNew,omitempty"`
+		ScriptCrawlerConfigJSON       string           `json:"scriptCrawlerConfigJson,omitempty"`
 		LastCrawlAt                   int64            `json:"lastCrawlAt,omitempty"`
 		GoogleDriveUseOnlineAPI       *bool            `json:"googleDriveUseOnlineAPI,omitempty"`
 		ScanGenerationStatus          GenerationStatus `json:"scanGenerationStatus"`
@@ -443,6 +465,9 @@ func (a *AdminServer) handleListDrives(w http.ResponseWriter, r *http.Request) {
 	}
 	list := make([]out, 0, len(drives))
 	for _, d := range drives {
+		if isCrawlerDriveKind(d.Kind) {
+			continue
+		}
 		counts := teaserCounts[d.ID]
 		thumbCounts := thumbnailCounts[d.ID]
 		fingerprintCount := fingerprintCounts[d.ID]
@@ -488,6 +513,11 @@ func (a *AdminServer) handleListDrives(w http.ResponseWriter, r *http.Request) {
 			TeaserEnabled:                 d.TeaserEnabled,
 			SkipDirIDs:                    append([]string{}, d.SkipDirIDs...),
 			Spider91Proxy:                 spider91ProxyForDrive(d),
+			ScriptCrawlerPythonPath:       scriptCrawlerCred(d, "python_path"),
+			ScriptCrawlerScriptPath:       scriptCrawlerCred(d, "script_path"),
+			ScriptCrawlerProxy:            scriptCrawlerCred(d, "proxy"),
+			ScriptCrawlerTargetNew:        scriptCrawlerCred(d, "target_new"),
+			ScriptCrawlerConfigJSON:       scriptCrawlerCred(d, "config_json"),
 			LastCrawlAt:                   lastCrawlAt,
 			GoogleDriveUseOnlineAPI:       googleDriveUseOnlineAPIForDrive(d),
 			ScanGenerationStatus:          generation.Scan,
@@ -543,7 +573,10 @@ func (a *AdminServer) handleUpsertDrive(w http.ResponseWriter, r *http.Request)
 		existing = existingDrive
 	}
 	if body.Kind == "spider91" {
-		credentials, err := mergeSpider91Credentials(existing, body.Credentials)
+		http.Error(w, "91Spider 已不再支持通过网盘添加，请在爬虫管理页面添加爬虫脚本", http.StatusBadRequest)
+		return
+	} else if body.Kind == scriptcrawler.Kind {
+		credentials, err := mergeScriptCrawlerCredentials(existing, body.Credentials)
 		if err != nil {
 			http.Error(w, err.Error(), http.StatusBadRequest)
 			return
@@ -600,6 +633,421 @@ func (a *AdminServer) handleUpsertDrive(w http.ResponseWriter, r *http.Request)
 	writeJSON(w, http.StatusOK, map[string]any{"ok": true})
 }

+type crawlerDTO struct {
+	ID                          string           `json:"id"`
+	Name                        string           `json:"name"`
+	Kind                        string           `json:"kind"`
+	Builtin                     string           `json:"builtin,omitempty"`
+	Status                      string           `json:"status"`
+	LastError                   string           `json:"lastError,omitempty"`
+	ScriptPath                  string           `json:"scriptPath"`
+	PythonPath                  string           `json:"pythonPath,omitempty"`
+	Proxy                       string           `json:"proxy,omitempty"`
+	TargetNew                   string           `json:"targetNew,omitempty"`
+	ConfigJSON                  string           `json:"configJson,omitempty"`
+	LastCrawlAt                 int64            `json:"lastCrawlAt,omitempty"`
+	ScanGenerationStatus        GenerationStatus `json:"scanGenerationStatus"`
+	ThumbnailGenerationStatus   GenerationStatus `json:"thumbnailGenerationStatus"`
+	PreviewGenerationStatus     GenerationStatus `json:"previewGenerationStatus"`
+	FingerprintGenerationStatus GenerationStatus `json:"fingerprintGenerationStatus"`
+	ThumbnailReadyCount         int              `json:"thumbnailReadyCount"`
+	ThumbnailPendingCount       int              `json:"thumbnailPendingCount"`
+	ThumbnailFailedCount        int              `json:"thumbnailFailedCount"`
+	TeaserReadyCount            int              `json:"teaserReadyCount"`
+	TeaserPendingCount          int              `json:"teaserPendingCount"`
+	TeaserFailedCount           int              `json:"teaserFailedCount"`
+	FingerprintReadyCount       int              `json:"fingerprintReadyCount"`
+	FingerprintPendingCount     int              `json:"fingerprintPendingCount"`
+	FingerprintFailedCount      int              `json:"fingerprintFailedCount"`
+}
+
+type upsertCrawlerReq struct {
+	ID         string `json:"id"`
+	Name       string `json:"name"`
+	Builtin    string `json:"builtin"`
+	ScriptPath string `json:"scriptPath"`
+	PythonPath string `json:"pythonPath"`
+	Proxy      string `json:"proxy"`
+	TargetNew  string `json:"targetNew"`
+	ConfigJSON string `json:"configJson"`
+}
+
+func (a *AdminServer) handleListCrawlers(w http.ResponseWriter, r *http.Request) {
+	all, err := a.Catalog.ListDrives(r.Context())
+	if err != nil {
+		writeErr(w, http.StatusInternalServerError, err)
+		return
+	}
+	teaserCounts, err := a.Catalog.CountTeasersByDrive(r.Context())
+	if err != nil {
+		writeErr(w, http.StatusInternalServerError, err)
+		return
+	}
+	thumbnailCounts, err := a.Catalog.CountThumbnailsByDrive(r.Context())
+	if err != nil {
+		writeErr(w, http.StatusInternalServerError, err)
+		return
+	}
+	fingerprintCounts, err := a.Catalog.CountFingerprintsByDrive(r.Context())
+	if err != nil {
+		writeErr(w, http.StatusInternalServerError, err)
+		return
+	}
+	generationStatuses := map[string]DriveGenerationStatuses{}
+	if a.GetDriveGenerationStatuses != nil {
+		generationStatuses = a.GetDriveGenerationStatuses()
+	}
+
+	out := []crawlerDTO{}
+	for _, d := range all {
+		if d == nil || !isCrawlerDriveKind(d.Kind) {
+			continue
+		}
+		out = append(out, a.crawlerDTOForDrive(d, teaserCounts[d.ID], thumbnailCounts[d.ID], fingerprintCounts[d.ID], generationStatuses[d.ID]))
+	}
+	writeJSON(w, http.StatusOK, out)
+}
+
+func (a *AdminServer) crawlerDTOForDrive(d *catalog.Drive, teaser catalog.DriveTeaserCounts, thumb catalog.DriveThumbnailCounts, fp catalog.DriveFingerprintCounts, generation DriveGenerationStatuses) crawlerDTO {
+	if generation.Scan.State == "" {
+		generation.Scan.State = "idle"
+	}
+	if generation.Thumbnail.State == "" {
+		generation.Thumbnail.State = "idle"
+	}
+	if generation.Preview.State == "" {
+		generation.Preview.State = "idle"
+	}
+	if generation.Fingerprint.State == "" {
+		generation.Fingerprint.State = "idle"
+	}
+	lastCrawlAt := int64(0)
+	if raw := strings.TrimSpace(d.Credentials["last_crawl_at"]); raw != "" {
+		if v, err := strconv.ParseInt(raw, 10, 64); err == nil {
+			lastCrawlAt = v
+		}
+	}
+	return crawlerDTO{
+		ID:                          d.ID,
+		Name:                        d.Name,
+		Kind:                        d.Kind,
+		Builtin:                     crawlerBuiltinForDrive(d),
+		Status:                      d.Status,
+		LastError:                   d.LastError,
+		ScriptPath:                  strings.TrimSpace(d.Credentials["script_path"]),
+		PythonPath:                  strings.TrimSpace(d.Credentials["python_path"]),
+		Proxy:                       strings.TrimSpace(d.Credentials["proxy"]),
+		TargetNew:                   strings.TrimSpace(d.Credentials["target_new"]),
+		ConfigJSON:                  strings.TrimSpace(d.Credentials["config_json"]),
+		LastCrawlAt:                 lastCrawlAt,
+		ScanGenerationStatus:        generation.Scan,
+		ThumbnailGenerationStatus:   generation.Thumbnail,
+		PreviewGenerationStatus:     generation.Preview,
+		FingerprintGenerationStatus: generation.Fingerprint,
+		ThumbnailReadyCount:         thumb.Ready,
+		ThumbnailPendingCount:       thumb.Pending,
+		ThumbnailFailedCount:        thumb.Failed,
+		TeaserReadyCount:            teaser.Ready,
+		TeaserPendingCount:          teaser.Pending,
+		TeaserFailedCount:           teaser.Failed,
+		FingerprintReadyCount:       fp.Ready,
+		FingerprintPendingCount:     fp.Pending,
+		FingerprintFailedCount:      fp.Failed,
+	}
+}
+
+func crawlerBuiltinForDrive(d *catalog.Drive) string {
+	if d == nil {
+		return ""
+	}
+	return strings.TrimSpace(d.Credentials["builtin"])
+}
+
+func (a *AdminServer) handleUpsertCrawler(w http.ResponseWriter, r *http.Request) {
+	var body upsertCrawlerReq
+	if err := json.NewDecoder(r.Body).Decode(&body); err != nil {
+		writeErr(w, http.StatusBadRequest, err)
+		return
+	}
+	id := strings.TrimSpace(body.ID)
+	name := strings.TrimSpace(body.Name)
+	if id == "" || name == "" {
+		http.Error(w, "id and name are required", http.StatusBadRequest)
+		return
+	}
+	existing, _ := a.Catalog.GetDrive(r.Context(), id)
+	creds := map[string]string{}
+	if existing != nil {
+		for k, v := range existing.Credentials {
+			creds[k] = v
+		}
+	}
+	builtin := strings.TrimSpace(body.Builtin)
+	if builtin != "" {
+		creds["builtin"] = builtin
+	}
+	scriptPath := strings.TrimSpace(body.ScriptPath)
+	if scriptPath == "" && builtin == "spider91" && a.DefaultSpider91ScriptPath != nil {
+		scriptPath = strings.TrimSpace(a.DefaultSpider91ScriptPath())
+	}
+	incoming := map[string]string{
+		"script_path": scriptPath,
+		"python_path": strings.TrimSpace(body.PythonPath),
+		"proxy":       strings.TrimSpace(body.Proxy),
+		"target_new":  strings.TrimSpace(body.TargetNew),
+		"config_json": strings.TrimSpace(body.ConfigJSON),
+	}
+	for k, v := range incoming {
+		creds[k] = v
+	}
+	merged, err := mergeScriptCrawlerCredentials(existing, creds)
+	if err != nil {
+		http.Error(w, err.Error(), http.StatusBadRequest)
+		return
+	}
+	if builtin != "" {
+		merged["builtin"] = builtin
+	}
+	d := &catalog.Drive{
+		ID:            id,
+		Kind:          scriptcrawler.Kind,
+		Name:          name,
+		RootID:        "/",
+		Credentials:   merged,
+		Status:        "disconnected",
+		TeaserEnabled: true,
+	}
+	if existing != nil {
+		d.TeaserEnabled = existing.TeaserEnabled
+	}
+	if err := a.Catalog.UpsertDrive(r.Context(), d); err != nil {
+		writeErr(w, http.StatusInternalServerError, err)
+		return
+	}
+	if a.OnDriveSaved != nil {
+		if err := a.OnDriveSaved(id); err != nil {
+			writeJSON(w, http.StatusOK, map[string]any{"ok": true, "warning": err.Error()})
+			return
+		}
+	}
+	writeJSON(w, http.StatusOK, map[string]any{"ok": true})
+}
+
+type importCrawlerScriptURLReq struct {
+	URL      string `json:"url"`
+	FileName string `json:"fileName"`
+}
+
+func (a *AdminServer) handleImportCrawlerScriptFile(w http.ResponseWriter, r *http.Request) {
+	r.Body = http.MaxBytesReader(w, r.Body, maxCrawlerScriptBytes+1024*1024)
+	if err := r.ParseMultipartForm(maxCrawlerScriptBytes + 1024*1024); err != nil {
+		writeErr(w, http.StatusBadRequest, err)
+		return
+	}
+	file, header, err := r.FormFile("file")
+	if err != nil {
+		writeErr(w, http.StatusBadRequest, errors.New("file is required"))
+		return
+	}
+	defer file.Close()
+
+	name := "crawler.py"
+	if header != nil && strings.TrimSpace(header.Filename) != "" {
+		name = header.Filename
+	}
+	scriptPath, err := a.saveCrawlerScript(r.Context(), name, file, maxCrawlerScriptBytes)
+	if err != nil {
+		writeErr(w, http.StatusBadRequest, err)
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]any{"scriptPath": scriptPath})
+}
+
+func (a *AdminServer) handleImportCrawlerScriptURL(w http.ResponseWriter, r *http.Request) {
+	var body importCrawlerScriptURLReq
+	if err := json.NewDecoder(r.Body).Decode(&body); err != nil {
+		writeErr(w, http.StatusBadRequest, err)
+		return
+	}
+	rawURL := strings.TrimSpace(body.URL)
+	u, err := url.Parse(rawURL)
+	if err != nil || u.Scheme == "" || u.Host == "" {
+		writeErr(w, http.StatusBadRequest, errors.New("脚本链接格式无效"))
+		return
+	}
+	if u.Scheme != "http" && u.Scheme != "https" {
+		writeErr(w, http.StatusBadRequest, errors.New("脚本链接仅支持 http:// 或 https://"))
+		return
+	}
+
+	client := a.HTTPClient
+	if client == nil {
+		client = &http.Client{Timeout: 30 * time.Second}
+	}
+	req, err := http.NewRequestWithContext(r.Context(), http.MethodGet, u.String(), nil)
+	if err != nil {
+		writeErr(w, http.StatusBadRequest, err)
+		return
+	}
+	req.Header.Set("User-Agent", "video-site-crawler-import/1.0")
+	resp, err := client.Do(req)
+	if err != nil {
+		writeErr(w, http.StatusBadGateway, err)
+		return
+	}
+	defer resp.Body.Close()
+	if resp.StatusCode < 200 || resp.StatusCode >= 300 {
+		writeErr(w, http.StatusBadGateway, fmt.Errorf("下载脚本失败: HTTP %d", resp.StatusCode))
+		return
+	}
+	if resp.ContentLength > maxCrawlerScriptBytes {
+		writeErr(w, http.StatusBadRequest, fmt.Errorf("脚本文件不能超过 %d KiB", maxCrawlerScriptBytes/1024))
+		return
+	}
+
+	name := strings.TrimSpace(body.FileName)
+	if name == "" {
+		name = path.Base(u.Path)
+	}
+	if name == "." || name == "/" || name == "" {
+		name = "crawler.py"
+	}
+	scriptPath, err := a.saveCrawlerScript(r.Context(), name, resp.Body, maxCrawlerScriptBytes)
+	if err != nil {
+		writeErr(w, http.StatusBadRequest, err)
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]any{"scriptPath": scriptPath})
+}
+
+func (a *AdminServer) saveCrawlerScript(ctx context.Context, name string, r io.Reader, maxBytes int64) (string, error) {
+	if err := ctx.Err(); err != nil {
+		return "", err
+	}
+	fileName, err := safeCrawlerScriptFileName(name)
+	if err != nil {
+		return "", err
+	}
+	root, err := a.crawlerScriptImportDir()
+	if err != nil {
+		return "", err
+	}
+	if err := os.MkdirAll(root, 0o755); err != nil {
+		return "", err
+	}
+	dst := filepath.Join(root, time.Now().UTC().Format("20060102T150405.000000000Z")+"-"+fileName)
+	dstAbs, err := filepath.Abs(dst)
+	if err != nil {
+		return "", err
+	}
+	rootAbs, err := filepath.Abs(root)
+	if err != nil {
+		return "", err
+	}
+	if dstAbs != rootAbs && !strings.HasPrefix(dstAbs, rootAbs+string(os.PathSeparator)) {
+		return "", errors.New("invalid crawler script path")
+	}
+
+	tmp := dstAbs + ".part"
+	out, err := os.OpenFile(tmp, os.O_CREATE|os.O_WRONLY|os.O_TRUNC, 0o644)
+	if err != nil {
+		return "", err
+	}
+	limited := io.LimitReader(r, maxBytes+1)
+	written, copyErr := io.Copy(out, limited)
+	closeErr := out.Close()
+	if copyErr != nil {
+		_ = os.Remove(tmp)
+		return "", copyErr
+	}
+	if closeErr != nil {
+		_ = os.Remove(tmp)
+		return "", closeErr
+	}
+	if written <= 0 {
+		_ = os.Remove(tmp)
+		return "", errors.New("脚本文件为空")
+	}
+	if written > maxBytes {
+		_ = os.Remove(tmp)
+		return "", fmt.Errorf("脚本文件不能超过 %d KiB", maxBytes/1024)
+	}
+	if err := os.Rename(tmp, dstAbs); err != nil {
+		_ = os.Remove(tmp)
+		return "", err
+	}
+	return dstAbs, nil
+}
+
+func (a *AdminServer) crawlerScriptImportDir() (string, error) {
+	base := strings.TrimSpace(a.LocalPreviewDir)
+	if base == "" {
+		base = filepath.Join(".", "data", "previews")
+	}
+	root := filepath.Join(filepath.Dir(base), "crawler-scripts")
+	return filepath.Abs(root)
+}
+
+func safeCrawlerScriptFileName(raw string) (string, error) {
+	name := strings.TrimSpace(filepath.Base(raw))
+	if name == "" || name == "." || name == string(os.PathSeparator) {
+		name = "crawler.py"
+	}
+	ext := strings.ToLower(filepath.Ext(name))
+	if ext != ".py" {
+		return "", errors.New("目前只支持导入 .py 爬虫脚本")
+	}
+	stem := strings.TrimSuffix(name, filepath.Ext(name))
+	var b strings.Builder
+	for _, r := range stem {
+		if (r >= 'a' && r <= 'z') || (r >= 'A' && r <= 'Z') || (r >= '0' && r <= '9') || r == '-' || r == '_' || r == '.' {
+			b.WriteRune(r)
+		} else {
+			b.WriteByte('_')
+		}
+	}
+	cleanStem := strings.Trim(b.String(), "._-")
+	if cleanStem == "" {
+		cleanStem = "crawler"
+	}
+	return cleanStem + ".py", nil
+}
+
+func (a *AdminServer) handleRunCrawler(w http.ResponseWriter, r *http.Request) {
+	id := chi.URLParam(r, "id")
+	status := a.nightlyJobStatus()
+	if status.Running || status.Queued {
+		writeJSON(w, http.StatusAccepted, map[string]any{
+			"ok":       true,
+			"accepted": false,
+			"message":  fullScanBusyMessage,
+			"status":   status,
+		})
+		return
+	}
+	accepted := true
+	if a.OnScanRequested != nil {
+		accepted = a.OnScanRequested(id)
+	}
+	resp := map[string]any{"ok": true, "accepted": accepted}
+	if !accepted {
+		resp["message"] = driveTaskBusyMessage
+	}
+	writeJSON(w, http.StatusAccepted, resp)
+}
+
+func (a *AdminServer) handleStopCrawlerTasks(w http.ResponseWriter, r *http.Request) {
+	a.handleStopDriveTasks(w, r)
+}
+
+func (a *AdminServer) handleDeleteCrawler(w http.ResponseWriter, r *http.Request) {
+	a.handleDeleteDrive(w, r)
+}
+
+func isCrawlerDriveKind(kind string) bool {
+	return kind == scriptcrawler.Kind
+}
+
 func spider91ProxyForDrive(d *catalog.Drive) string {
 	if d == nil || d.Kind != "spider91" || d.Credentials == nil {
 		return ""
@@ -607,6 +1055,13 @@ func spider91ProxyForDrive(d *catalog.Drive) string {
 	return strings.TrimSpace(d.Credentials["proxy"])
 }

+func scriptCrawlerCred(d *catalog.Drive, key string) string {
+	if d == nil || d.Kind != scriptcrawler.Kind || d.Credentials == nil {
+		return ""
+	}
+	return strings.TrimSpace(d.Credentials[key])
+}
+
 func googleDriveUseOnlineAPIForDrive(d *catalog.Drive) *bool {
 	if d == nil || d.Kind != "googledrive" {
 		return nil
@@ -676,20 +1131,89 @@ func mergeSpider91Credentials(existing *catalog.Drive, incoming map[string]strin
 	return merged, nil
 }

+func mergeScriptCrawlerCredentials(existing *catalog.Drive, incoming map[string]string) (map[string]string, error) {
+	merged := map[string]string{}
+	if existing != nil {
+		for k, v := range existing.Credentials {
+			merged[k] = v
+		}
+	}
+	for k, v := range incoming {
+		key := strings.TrimSpace(k)
+		if key == "" {
+			continue
+		}
+		value := strings.TrimSpace(v)
+		switch key {
+		case "proxy":
+			proxy, err := normalizeCrawlerProxyURL(value, "脚本爬虫")
+			if err != nil {
+				return nil, err
+			}
+			if proxy == "" {
+				delete(merged, key)
+			} else {
+				merged[key] = proxy
+			}
+		case "target_new":
+			if value == "" {
+				delete(merged, key)
+				continue
+			}
+			n, err := strconv.Atoi(value)
+			if err != nil || n <= 0 {
+				return nil, fmt.Errorf("脚本爬虫 target_new 必须是正整数")
+			}
+			merged[key] = strconv.Itoa(n)
+		case "config_json":
+			if value == "" {
+				delete(merged, key)
+				continue
+			}
+			if !json.Valid([]byte(value)) {
+				return nil, fmt.Errorf("脚本爬虫自定义配置必须是合法 JSON")
+			}
+			merged[key] = value
+		case "python_path", "script_path":
+			if value == "" {
+				if existing == nil || key == "script_path" {
+					delete(merged, key)
+				}
+				continue
+			}
+			merged[key] = value
+		default:
+			if value == "" {
+				delete(merged, key)
+			} else {
+				merged[key] = value
+			}
+		}
+	}
+	if strings.TrimSpace(merged["script_path"]) == "" && !strings.EqualFold(strings.TrimSpace(merged["builtin"]), "spider91") {
+		return nil, fmt.Errorf("脚本爬虫必须填写 script_path")
+	}
+	return merged, nil
+}
+
 func normalizeSpider91ProxyURL(raw string) (string, error) {
+	return normalizeCrawlerProxyURL(raw, "91Spider")
+}
+
+func normalizeCrawlerProxyURL(raw, label string) (string, error) {
 	proxy := strings.TrimSpace(raw)
 	if proxy == "" {
 		return "", nil
 	}
 	u, err := url.Parse(proxy)
 	if err != nil || u.Scheme == "" || u.Host == "" {
-		return "", fmt.Errorf("91Spider 代理地址格式无效，请填写类似 http://127.0.0.1:7890 的地址")
+		return "", fmt.Errorf("%s 代理地址格式无效，请填写类似 http://127.0.0.1:7890 的地址", label)
 	}
 	switch strings.ToLower(u.Scheme) {
 	case "http", "https", "socks5", "socks5h":
 		return proxy, nil
 	default:
-		return "", fmt.Errorf("91Spider 代理地址仅支持 http://、https://、socks5:// 或 socks5h://")
+		return "", fmt.Errorf("%s 代理地址仅支持 http://、https://、socks5:// 或 socks5h://", label)
 	}
 }

@@ -5,6 +5,7 @@ import (
 	"context"
 	"database/sql"
 	"encoding/json"
+	"mime/multipart"
 	"net/http"
 	"net/http/httptest"
 	"os"
@@ -672,7 +673,7 @@ func TestHandleUpsertGoogleDriveMergesOAuthCredentials(t *testing.T) {
 	}
 }

-func TestHandleUpsertSpider91ProxyPreservesRuntimeCredentials(t *testing.T) {
+func TestHandleUpsertSpider91DriveIsRejected(t *testing.T) {
 	ctx := context.Background()
 	cat, err := catalog.Open(t.TempDir() + "/catalog.db")
 	if err != nil {
@@ -708,16 +709,19 @@ func TestHandleUpsertSpider91ProxyPreservesRuntimeCredentials(t *testing.T) {
 	}`))
 	rr := httptest.NewRecorder()
 	(&AdminServer{Catalog: cat}).handleUpsertDrive(rr, req)
-	if rr.Code != http.StatusOK {
-		t.Fatalf("status = %d, body = %s", rr.Code, rr.Body.String())
+	if rr.Code != http.StatusBadRequest {
+		t.Fatalf("status = %d, want 400; body = %s", rr.Code, rr.Body.String())
+	}
+	if !strings.Contains(rr.Body.String(), "爬虫管理") {
+		t.Fatalf("body = %q, want crawler management guidance", rr.Body.String())
 	}

 	got, err := cat.GetDrive(ctx, "spider91-main")
 	if err != nil {
 		t.Fatalf("get drive: %v", err)
 	}
-	if got.Credentials["proxy"] != "socks5h://proxy-user:proxy-pass@127.0.0.1:7891" {
-		t.Fatalf("proxy = %q, want trimmed new proxy", got.Credentials["proxy"])
+	if got.Credentials["proxy"] != "http://old-proxy.local:7890" {
+		t.Fatalf("proxy = %q, want unchanged old proxy", got.Credentials["proxy"])
 	}
 	if got.Credentials["last_crawl_at"] != "1800000000" {
 		t.Fatalf("last_crawl_at = %q, want preserved", got.Credentials["last_crawl_at"])
@@ -725,59 +729,6 @@ func TestHandleUpsertSpider91ProxyPreservesRuntimeCredentials(t *testing.T) {
 	if got.Credentials["script_path"] == "" {
 		t.Fatalf("script_path should be preserved")
 	}
-
-	req = httptest.NewRequest(http.MethodPost, "/admin/api/drives", strings.NewReader(`{
-		"id": "spider91-main",
-		"kind": "spider91",
-		"name": "91 Spider",
-		"rootId": "/",
-		"credentials": {"proxy": " "}
-	}`))
-	rr = httptest.NewRecorder()
-	(&AdminServer{Catalog: cat}).handleUpsertDrive(rr, req)
-	if rr.Code != http.StatusOK {
-		t.Fatalf("clear status = %d, body = %s", rr.Code, rr.Body.String())
-	}
-
-	got, err = cat.GetDrive(ctx, "spider91-main")
-	if err != nil {
-		t.Fatalf("get cleared drive: %v", err)
-	}
-	if _, ok := got.Credentials["proxy"]; ok {
-		t.Fatalf("proxy should be removed after empty save, got %q", got.Credentials["proxy"])
-	}
-	if got.Credentials["last_crawl_at"] != "1800000000" {
-		t.Fatalf("last_crawl_at after clear = %q, want preserved", got.Credentials["last_crawl_at"])
-	}
-}
-
-func TestHandleUpsertSpider91RejectsUnsupportedProxyScheme(t *testing.T) {
-	cat, err := catalog.Open(t.TempDir() + "/catalog.db")
-	if err != nil {
-		t.Fatalf("open catalog: %v", err)
-	}
-	t.Cleanup(func() {
-		if err := cat.Close(); err != nil {
-			t.Fatalf("close catalog: %v", err)
-		}
-	})
-
-	req := httptest.NewRequest(http.MethodPost, "/admin/api/drives", strings.NewReader(`{
-		"id": "spider91-main",
-		"kind": "spider91",
-		"name": "91 Spider",
-		"rootId": "/",
-		"credentials": {"proxy": "ftp://127.0.0.1:21"}
-	}`))
-	rr := httptest.NewRecorder()
-	(&AdminServer{Catalog: cat}).handleUpsertDrive(rr, req)
-
-	if rr.Code != http.StatusBadRequest {
-		t.Fatalf("status = %d, want 400; body = %s", rr.Code, rr.Body.String())
-	}
-	if !strings.Contains(rr.Body.String(), "socks5:// 或 socks5h://") {
-		t.Fatalf("body = %q, want supported schemes message", rr.Body.String())
-	}
 }

 func TestHandleDeleteDriveRunsRequestedCleanupBeforeDeletingDrive(t *testing.T) {
@@ -890,7 +841,7 @@ func TestHandleDeleteDriveRequiresCleanupConfirmation(t *testing.T) {
 	}
 }

-func TestHandleListDrivesIncludesSpider91Proxy(t *testing.T) {
+func TestHandleListCrawlersOnlyIncludesCrawlerPageScripts(t *testing.T) {
 	ctx := context.Background()
 	cat, err := catalog.Open(t.TempDir() + "/catalog.db")
 	if err != nil {
@@ -911,6 +862,20 @@ func TestHandleListDrivesIncludesSpider91Proxy(t *testing.T) {
 			Credentials: map[string]string{
 				"last_crawl_at": "1800000000",
 				"proxy":         " http://127.0.0.1:7890 ",
+				"script_path":   "/opt/video-site-91/91VideoSpider/spider_91porn.py",
+			},
+			Status: "ok",
+		},
+		{
+			ID:     "crawler-spider91",
+			Kind:   "scriptcrawler",
+			Name:   "91 Spider",
+			RootID: "/",
+			Credentials: map[string]string{
+				"builtin":       "spider91",
+				"last_crawl_at": "1800000000",
+				"proxy":         " http://127.0.0.1:7890 ",
+				"script_path":   "/opt/video-site-91/91VideoSpider/spider_91porn.py",
 			},
 			Status: "ok",
 		},
@@ -930,39 +895,251 @@ func TestHandleListDrivesIncludesSpider91Proxy(t *testing.T) {
 		}
 	}

-	req := httptest.NewRequest(http.MethodGet, "/admin/api/drives", nil)
+	req := httptest.NewRequest(http.MethodGet, "/admin/api/crawlers", nil)
 	rr := httptest.NewRecorder()
-	(&AdminServer{Catalog: cat}).handleListDrives(rr, req)
+	srv := &AdminServer{Catalog: cat}
+	srv.handleListCrawlers(rr, req)
 	if rr.Code != http.StatusOK {
 		t.Fatalf("status = %d, body = %s", rr.Code, rr.Body.String())
 	}

 	var got []struct {
-		ID            string `json:"id"`
-		Spider91Proxy string `json:"spider91Proxy"`
-		LastCrawlAt   int64  `json:"lastCrawlAt"`
+		ID          string `json:"id"`
+		Kind        string `json:"kind"`
+		Builtin     string `json:"builtin"`
+		Proxy       string `json:"proxy"`
+		LastCrawlAt int64  `json:"lastCrawlAt"`
 	}
 	if err := json.NewDecoder(rr.Body).Decode(&got); err != nil {
 		t.Fatalf("decode: %v", err)
 	}
 	byID := map[string]struct {
-		Spider91Proxy string
-		LastCrawlAt   int64
+		Kind        string
+		Builtin     string
+		Proxy       string
+		LastCrawlAt int64
 	}{}
 	for _, d := range got {
 		byID[d.ID] = struct {
-			Spider91Proxy string
-			LastCrawlAt   int64
-		}{Spider91Proxy: d.Spider91Proxy, LastCrawlAt: d.LastCrawlAt}
+			Kind        string
+			Builtin     string
+			Proxy       string
+			LastCrawlAt int64
+		}{Kind: d.Kind, Builtin: d.Builtin, Proxy: d.Proxy, LastCrawlAt: d.LastCrawlAt}
 	}
-	if byID["spider91-main"].Spider91Proxy != "http://127.0.0.1:7890" {
-		t.Fatalf("spider91 proxy = %q, want trimmed proxy", byID["spider91-main"].Spider91Proxy)
+	if _, ok := byID["spider91-main"]; ok {
+		t.Fatal("legacy spider91 drive should not be returned by crawler list")
 	}
-	if byID["spider91-main"].LastCrawlAt != 1800000000 {
-		t.Fatalf("lastCrawlAt = %d, want 1800000000", byID["spider91-main"].LastCrawlAt)
+	if byID["crawler-spider91"].Kind != "scriptcrawler" || byID["crawler-spider91"].Builtin != "spider91" {
+		t.Fatalf("crawler kind/builtin = %q/%q, want scriptcrawler/spider91", byID["crawler-spider91"].Kind, byID["crawler-spider91"].Builtin)
 	}
-	if byID["onedrive-main"].Spider91Proxy != "" {
-		t.Fatalf("onedrive spider91Proxy = %q, want empty", byID["onedrive-main"].Spider91Proxy)
+	if byID["crawler-spider91"].Proxy != "http://127.0.0.1:7890" {
+		t.Fatalf("crawler proxy = %q, want trimmed proxy", byID["crawler-spider91"].Proxy)
+	}
+	if byID["crawler-spider91"].LastCrawlAt != 1800000000 {
+		t.Fatalf("lastCrawlAt = %d, want 1800000000", byID["crawler-spider91"].LastCrawlAt)
+	}
+	if _, ok := byID["onedrive-main"]; ok {
+		t.Fatal("onedrive should not be returned by crawler list")
+	}
+
+	driveReq := httptest.NewRequest(http.MethodGet, "/admin/api/drives", nil)
+	driveRR := httptest.NewRecorder()
+	srv.handleListDrives(driveRR, driveReq)
+	if driveRR.Code != http.StatusOK {
+		t.Fatalf("drive status = %d, body = %s", driveRR.Code, driveRR.Body.String())
+	}
+	var drives []struct {
+		ID string `json:"id"`
+	}
+	if err := json.NewDecoder(driveRR.Body).Decode(&drives); err != nil {
+		t.Fatalf("decode drives: %v", err)
+	}
+	driveIDs := map[string]bool{}
+	for _, d := range drives {
+		driveIDs[d.ID] = true
+	}
+	if !driveIDs["spider91-main"] {
+		t.Fatal("legacy spider91 drive should remain visible in drive list for deletion")
+	}
+	if driveIDs["crawler-spider91"] {
+		t.Fatal("scriptcrawler should not be returned by drive list")
+	}
+}
+
+func TestHandleUpsertCrawlerAllowsBuiltinSpider91WithoutScriptPath(t *testing.T) {
+	ctx := context.Background()
+	cat, err := catalog.Open(t.TempDir() + "/catalog.db")
+	if err != nil {
+		t.Fatalf("open catalog: %v", err)
+	}
+	t.Cleanup(func() {
+		if err := cat.Close(); err != nil {
+			t.Fatalf("close catalog: %v", err)
+		}
+	})
+
+	req := httptest.NewRequest(http.MethodPost, "/admin/api/crawlers", strings.NewReader(`{
+		"id": "spider91-main",
+		"name": "91 Spider",
+		"builtin": "spider91",
+		"scriptPath": "",
+		"pythonPath": "python3",
+		"targetNew": "15"
+	}`))
+	rr := httptest.NewRecorder()
+	(&AdminServer{
+		Catalog: cat,
+		DefaultSpider91ScriptPath: func() string {
+			return ""
+		},
+	}).handleUpsertCrawler(rr, req)
+	if rr.Code != http.StatusOK {
+		t.Fatalf("status = %d, body = %s", rr.Code, rr.Body.String())
+	}
+
+	got, err := cat.GetDrive(ctx, "spider91-main")
+	if err != nil {
+		t.Fatalf("get crawler drive: %v", err)
+	}
+	if got.Kind != "scriptcrawler" || got.Credentials["builtin"] != "spider91" {
+		t.Fatalf("kind/builtin = %q/%q, want scriptcrawler/spider91", got.Kind, got.Credentials["builtin"])
+	}
+	if got.Credentials["script_path"] != "" {
+		t.Fatalf("script_path = %q, want empty when default is unavailable", got.Credentials["script_path"])
+	}
+}
+
+func TestHandleImportCrawlerScriptFile(t *testing.T) {
+	tmp := t.TempDir()
+	var body bytes.Buffer
+	mw := multipart.NewWriter(&body)
+	part, err := mw.CreateFormFile("file", "../demo crawler.py")
+	if err != nil {
+		t.Fatalf("create form file: %v", err)
+	}
+	if _, err := part.Write([]byte("print('crawler')\n")); err != nil {
+		t.Fatalf("write part: %v", err)
+	}
+	if err := mw.Close(); err != nil {
+		t.Fatalf("close multipart: %v", err)
+	}
+
+	req := httptest.NewRequest(http.MethodPost, "/admin/api/crawlers/import-file", &body)
+	req.Header.Set("Content-Type", mw.FormDataContentType())
+	rr := httptest.NewRecorder()
+	(&AdminServer{LocalPreviewDir: filepath.Join(tmp, "previews")}).handleImportCrawlerScriptFile(rr, req)
+	if rr.Code != http.StatusOK {
+		t.Fatalf("status = %d, body = %s", rr.Code, rr.Body.String())
+	}
+	var got struct {
+		ScriptPath string `json:"scriptPath"`
+	}
+	if err := json.NewDecoder(rr.Body).Decode(&got); err != nil {
+		t.Fatalf("decode: %v", err)
+	}
+	wantRoot := filepath.Join(tmp, "crawler-scripts")
+	if !strings.HasPrefix(got.ScriptPath, wantRoot+string(os.PathSeparator)) {
+		t.Fatalf("script path = %q, want under %q", got.ScriptPath, wantRoot)
+	}
+	if filepath.Ext(got.ScriptPath) != ".py" {
+		t.Fatalf("script path = %q, want .py", got.ScriptPath)
+	}
+	data, err := os.ReadFile(got.ScriptPath)
+	if err != nil {
+		t.Fatalf("read imported script: %v", err)
+	}
+	if string(data) != "print('crawler')\n" {
+		t.Fatalf("script content = %q", string(data))
+	}
+}
+
+func TestHandleImportCrawlerScriptFileRejectsNonPython(t *testing.T) {
+	tmp := t.TempDir()
+	var body bytes.Buffer
+	mw := multipart.NewWriter(&body)
+	part, err := mw.CreateFormFile("file", "crawler.txt")
+	if err != nil {
+		t.Fatalf("create form file: %v", err)
+	}
+	if _, err := part.Write([]byte("print('crawler')\n")); err != nil {
+		t.Fatalf("write part: %v", err)
+	}
+	if err := mw.Close(); err != nil {
+		t.Fatalf("close multipart: %v", err)
+	}
+
+	req := httptest.NewRequest(http.MethodPost, "/admin/api/crawlers/import-file", &body)
+	req.Header.Set("Content-Type", mw.FormDataContentType())
+	rr := httptest.NewRecorder()
+	(&AdminServer{LocalPreviewDir: filepath.Join(tmp, "previews")}).handleImportCrawlerScriptFile(rr, req)
+	if rr.Code != http.StatusBadRequest {
+		t.Fatalf("status = %d, want 400; body = %s", rr.Code, rr.Body.String())
+	}
+	if !strings.Contains(rr.Body.String(), ".py") {
+		t.Fatalf("body = %s, want .py error", rr.Body.String())
+	}
+}
+
+func TestHandleImportCrawlerScriptURL(t *testing.T) {
+	tmp := t.TempDir()
+	upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		if r.URL.Path != "/crawler.py" {
+			http.NotFound(w, r)
+			return
+		}
+		_, _ = w.Write([]byte("# crawler from url\n"))
+	}))
+	defer upstream.Close()
+
+	req := httptest.NewRequest(http.MethodPost, "/admin/api/crawlers/import-url", strings.NewReader(`{
+		"url": "`+upstream.URL+`/crawler.py"
+	}`))
+	rr := httptest.NewRecorder()
+	(&AdminServer{LocalPreviewDir: filepath.Join(tmp, "previews")}).handleImportCrawlerScriptURL(rr, req)
+	if rr.Code != http.StatusOK {
+		t.Fatalf("status = %d, body = %s", rr.Code, rr.Body.String())
+	}
+	var got struct {
+		ScriptPath string `json:"scriptPath"`
+	}
+	if err := json.NewDecoder(rr.Body).Decode(&got); err != nil {
+		t.Fatalf("decode: %v", err)
+	}
+	wantRoot := filepath.Join(tmp, "crawler-scripts")
+	if !strings.HasPrefix(got.ScriptPath, wantRoot+string(os.PathSeparator)) {
+		t.Fatalf("script path = %q, want under %q", got.ScriptPath, wantRoot)
+	}
+	data, err := os.ReadFile(got.ScriptPath)
+	if err != nil {
+		t.Fatalf("read imported script: %v", err)
+	}
+	if string(data) != "# crawler from url\n" {
+		t.Fatalf("script content = %q", string(data))
+	}
+}
+
+func TestHandleImportCrawlerScriptURLRejectsNonPython(t *testing.T) {
+	tmp := t.TempDir()
+	upstream := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		if r.URL.Path != "/crawler.txt" {
+			http.NotFound(w, r)
+			return
+		}
+		_, _ = w.Write([]byte("# crawler from url\n"))
+	}))
+	defer upstream.Close()
+
+	req := httptest.NewRequest(http.MethodPost, "/admin/api/crawlers/import-url", strings.NewReader(`{
+		"url": "`+upstream.URL+`/crawler.txt"
+	}`))
+	rr := httptest.NewRecorder()
+	(&AdminServer{LocalPreviewDir: filepath.Join(tmp, "previews")}).handleImportCrawlerScriptURL(rr, req)
+	if rr.Code != http.StatusBadRequest {
+		t.Fatalf("status = %d, want 400; body = %s", rr.Code, rr.Body.String())
+	}
+	if !strings.Contains(rr.Body.String(), ".py") {
+		t.Fatalf("body = %s, want .py error", rr.Body.String())
 	}
 }

@@ -706,7 +706,19 @@ func (c *Catalog) ListVideoFileIDsByDrive(ctx context.Context, driveID string) (
 // 用途：crawler 把这个集合写到 seen 文件，让 Python/Go 跳过已爬过的视频，
 // 配合 --target-new 真正凑出 N 个未爬过的视频。
 func (c *Catalog) ListSpider91Viewkeys(ctx context.Context, driveID string) ([]string, error) {
-	prefix := "spider91-" + driveID + "-"
+	return c.ListCrawlerSourceIDs(ctx, "spider91", driveID)
+}
+
+// ListCrawlerSourceIDs lists source IDs that were already imported by a
+// crawler-like drive. It reads both videos and deleted_videos so explicit admin
+// deletions remain tombstoned for future crawler runs.
+func (c *Catalog) ListCrawlerSourceIDs(ctx context.Context, kind, driveID string) ([]string, error) {
+	kind = strings.TrimSpace(kind)
+	driveID = strings.TrimSpace(driveID)
+	if kind == "" || driveID == "" {
+		return nil, nil
+	}
+	prefix := kind + "-" + driveID + "-"
 	rows, err := c.db.QueryContext(ctx,
 		`SELECT SUBSTR(id, ?) FROM videos WHERE id LIKE ? || '%'
 		 UNION
@@ -0,0 +1,268 @@
+package scriptcrawler
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"net/http"
+	"net/http/httptest"
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	"github.com/video-site/backend/internal/catalog"
+)
+
+func TestCrawlerRunOnceImportsLocalFileAndSkipsExisting(t *testing.T) {
+	ctx := context.Background()
+	tmp := t.TempDir()
+	cat, err := catalog.Open(filepath.Join(tmp, "catalog.db"))
+	if err != nil {
+		t.Fatalf("open catalog: %v", err)
+	}
+	t.Cleanup(func() {
+		if err := cat.Close(); err != nil {
+			t.Fatalf("close catalog: %v", err)
+		}
+	})
+	drv := New(Config{ID: "demo", RootDir: filepath.Join(tmp, "crawler")})
+	if err := drv.Init(ctx); err != nil {
+		t.Fatalf("driver init: %v", err)
+	}
+	dummyScript := filepath.Join(tmp, "helper-script")
+	if err := os.WriteFile(dummyScript, []byte("helper"), 0o755); err != nil {
+		t.Fatalf("write dummy script: %v", err)
+	}
+	wrapper := filepath.Join(tmp, "helper-wrapper.sh")
+	wrapperScript := fmt.Sprintf("#!/bin/sh\nexec %q -test.run=TestScriptCrawlerHelperProcess \"$@\"\n", os.Args[0])
+	if err := os.WriteFile(wrapper, []byte(wrapperScript), 0o755); err != nil {
+		t.Fatalf("write helper wrapper: %v", err)
+	}
+
+	t.Setenv("GO_WANT_SCRIPTCRAWLER_HELPER", "1")
+	c := NewCrawler(CrawlerConfig{
+		Driver:     drv,
+		Catalog:    cat,
+		PythonPath: wrapper,
+		ScriptPath: dummyScript,
+	})
+	res, err := c.RunOnce(ctx, 1)
+	if err != nil {
+		t.Fatalf("run once: %v", err)
+	}
+	if res.NewVideos != 1 || res.Skipped != 0 || res.Failed != 0 {
+		t.Fatalf("result = new:%d skipped:%d failed:%d, want 1/0/0", res.NewVideos, res.Skipped, res.Failed)
+	}
+	v, err := cat.GetVideo(ctx, BuildVideoID("demo", "abc-123"))
+	if err != nil {
+		t.Fatalf("get video: %v", err)
+	}
+	if v.Title != "Imported From Helper" || v.FileID != "abc-123.mp4" || v.Size == 0 {
+		t.Fatalf("video = title:%q file:%q size:%d", v.Title, v.FileID, v.Size)
+	}
+	if _, err := os.Stat(filepath.Join(drv.VideosDir(), "abc-123.mp4")); err != nil {
+		t.Fatalf("video file not copied: %v", err)
+	}
+
+	res, err = c.RunOnce(ctx, 1)
+	if err != nil {
+		t.Fatalf("second run: %v", err)
+	}
+	if res.NewVideos != 0 || res.Skipped != 1 {
+		t.Fatalf("second result = new:%d skipped:%d, want 0/1", res.NewVideos, res.Skipped)
+	}
+	if res.SeenSnapshot != 1 {
+		t.Fatalf("seen snapshot = %d, want 1", res.SeenSnapshot)
+	}
+}
+
+func TestCrawlerRunOnceUsesSourceKindNamespace(t *testing.T) {
+	ctx := context.Background()
+	tmp := t.TempDir()
+	cat, err := catalog.Open(filepath.Join(tmp, "catalog.db"))
+	if err != nil {
+		t.Fatalf("open catalog: %v", err)
+	}
+	t.Cleanup(func() {
+		if err := cat.Close(); err != nil {
+			t.Fatalf("close catalog: %v", err)
+		}
+	})
+	drv := New(Config{ID: "demo", RootDir: filepath.Join(tmp, "crawler")})
+	if err := drv.Init(ctx); err != nil {
+		t.Fatalf("driver init: %v", err)
+	}
+	dummyScript := filepath.Join(tmp, "helper-script")
+	if err := os.WriteFile(dummyScript, []byte("helper"), 0o755); err != nil {
+		t.Fatalf("write dummy script: %v", err)
+	}
+	wrapper := filepath.Join(tmp, "helper-wrapper.sh")
+	wrapperScript := fmt.Sprintf("#!/bin/sh\nexec %q -test.run=TestScriptCrawlerHelperProcess \"$@\"\n", os.Args[0])
+	if err := os.WriteFile(wrapper, []byte(wrapperScript), 0o755); err != nil {
+		t.Fatalf("write helper wrapper: %v", err)
+	}
+
+	t.Setenv("GO_WANT_SCRIPTCRAWLER_HELPER", "1")
+	c := NewCrawler(CrawlerConfig{
+		Driver:     drv,
+		Catalog:    cat,
+		SourceKind: "spider91",
+		PythonPath: wrapper,
+		ScriptPath: dummyScript,
+	})
+	res, err := c.RunOnce(ctx, 1)
+	if err != nil {
+		t.Fatalf("run once: %v", err)
+	}
+	if res.NewVideos != 1 || res.SeenSnapshot != 0 {
+		t.Fatalf("result = new:%d seen:%d, want 1/0", res.NewVideos, res.SeenSnapshot)
+	}
+	videoID := BuildVideoIDForKind("spider91", "demo", "abc-123")
+	if _, err := cat.GetVideo(ctx, videoID); err != nil {
+		t.Fatalf("get source-kind video: %v", err)
+	}
+	if _, err := cat.GetVideo(ctx, BuildVideoID("demo", "abc-123")); err == nil {
+		t.Fatalf("default namespace video unexpectedly exists")
+	}
+
+	res, err = c.RunOnce(ctx, 1)
+	if err != nil {
+		t.Fatalf("second run: %v", err)
+	}
+	if res.NewVideos != 0 || res.Skipped != 1 || res.SeenSnapshot != 1 {
+		t.Fatalf("second result = new:%d skipped:%d seen:%d, want 0/1/1", res.NewVideos, res.Skipped, res.SeenSnapshot)
+	}
+}
+
+func TestCrawlerRunOnceImportsSimpleMediaURLWithoutSourceID(t *testing.T) {
+	ctx := context.Background()
+	tmp := t.TempDir()
+	cat, err := catalog.Open(filepath.Join(tmp, "catalog.db"))
+	if err != nil {
+		t.Fatalf("open catalog: %v", err)
+	}
+	t.Cleanup(func() {
+		if err := cat.Close(); err != nil {
+			t.Fatalf("close catalog: %v", err)
+		}
+	})
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		if r.URL.Path != "/video.mp4" {
+			http.NotFound(w, r)
+			return
+		}
+		_, _ = w.Write([]byte("simple-video-bytes"))
+	}))
+	defer srv.Close()
+
+	drv := New(Config{ID: "demo", RootDir: filepath.Join(tmp, "crawler")})
+	if err := drv.Init(ctx); err != nil {
+		t.Fatalf("driver init: %v", err)
+	}
+	dummyScript := filepath.Join(tmp, "helper-script")
+	if err := os.WriteFile(dummyScript, []byte("helper"), 0o755); err != nil {
+		t.Fatalf("write dummy script: %v", err)
+	}
+	wrapper := filepath.Join(tmp, "helper-wrapper.sh")
+	wrapperScript := fmt.Sprintf("#!/bin/sh\nexec %q -test.run=TestScriptCrawlerHelperProcess \"$@\"\n", os.Args[0])
+	if err := os.WriteFile(wrapper, []byte(wrapperScript), 0o755); err != nil {
+		t.Fatalf("write helper wrapper: %v", err)
+	}
+
+	t.Setenv("GO_WANT_SCRIPTCRAWLER_HELPER", "1")
+	t.Setenv("GO_WANT_SCRIPTCRAWLER_SIMPLE", "1")
+	t.Setenv("GO_SCRIPTCRAWLER_MEDIA_URL", srv.URL+"/video.mp4?token=first")
+	c := NewCrawler(CrawlerConfig{
+		Driver:     drv,
+		Catalog:    cat,
+		PythonPath: wrapper,
+		ScriptPath: dummyScript,
+		HTTPClient: srv.Client(),
+	})
+	res, err := c.RunOnce(ctx, 1)
+	if err != nil {
+		t.Fatalf("run once: %v", err)
+	}
+	if res.NewVideos != 1 || res.Skipped != 0 || res.Failed != 0 {
+		t.Fatalf("result = new:%d skipped:%d failed:%d, want 1/0/0", res.NewVideos, res.Skipped, res.Failed)
+	}
+	videos, err := cat.ListVideosByDrive(ctx, "demo")
+	if err != nil {
+		t.Fatalf("list videos: %v", err)
+	}
+	if len(videos) != 1 {
+		t.Fatalf("videos = %d, want 1", len(videos))
+	}
+	v := videos[0]
+	if !strings.HasPrefix(v.ID, BuildVideoID("demo", "auto-")) {
+		t.Fatalf("video id = %q, want generated auto source id", v.ID)
+	}
+	if v.Title != "Simple Protocol Video" || v.Ext != "mp4" || v.ThumbnailURL != "" || v.Size == 0 {
+		t.Fatalf("video = title:%q ext:%q thumb:%q size:%d", v.Title, v.Ext, v.ThumbnailURL, v.Size)
+	}
+	if _, err := os.Stat(filepath.Join(drv.VideosDir(), v.FileID)); err != nil {
+		t.Fatalf("video file not downloaded: %v", err)
+	}
+
+	t.Setenv("GO_SCRIPTCRAWLER_MEDIA_URL", srv.URL+"/video.mp4?token=second")
+	res, err = c.RunOnce(ctx, 1)
+	if err != nil {
+		t.Fatalf("second run: %v", err)
+	}
+	if res.NewVideos != 0 || res.Skipped != 1 {
+		t.Fatalf("second result = new:%d skipped:%d, want 0/1", res.NewVideos, res.Skipped)
+	}
+}
+
+func TestScriptCrawlerHelperProcess(t *testing.T) {
+	if os.Getenv("GO_WANT_SCRIPTCRAWLER_HELPER") != "1" {
+		return
+	}
+	args := os.Args
+	jobPath := ""
+	for i := 0; i < len(args)-1; i++ {
+		if args[i] == "--job" {
+			jobPath = args[i+1]
+			break
+		}
+	}
+	if jobPath == "" {
+		fmt.Fprintln(os.Stderr, "missing --job")
+		os.Exit(2)
+	}
+	data, err := os.ReadFile(jobPath)
+	if err != nil {
+		fmt.Fprintln(os.Stderr, err)
+		os.Exit(2)
+	}
+	var job Job
+	if err := json.Unmarshal(data, &job); err != nil {
+		fmt.Fprintln(os.Stderr, err)
+		os.Exit(2)
+	}
+	if os.Getenv("GO_WANT_SCRIPTCRAWLER_SIMPLE") == "1" {
+		event := map[string]any{
+			"title":     "Simple Protocol Video",
+			"media_url": os.Getenv("GO_SCRIPTCRAWLER_MEDIA_URL"),
+		}
+		_ = json.NewEncoder(os.Stdout).Encode(event)
+		os.Exit(0)
+	}
+	localFile := filepath.Join(job.OutputDir, "helper.mp4")
+	if err := os.WriteFile(localFile, []byte("helper-video"), 0o644); err != nil {
+		fmt.Fprintln(os.Stderr, err)
+		os.Exit(2)
+	}
+	event := Event{
+		Type: "item",
+		Item: Item{
+			SourceID: "abc-123",
+			Title:    "Imported From Helper",
+			Author:   "helper",
+			Media:    MediaRef{LocalFile: localFile},
+		},
+	}
+	_ = json.NewEncoder(os.Stdout).Encode(event)
+	os.Exit(0)
+}
@@ -0,0 +1,172 @@
+// Package scriptcrawler provides a generic local drive for script-based
+// crawlers. A crawler script discovers videos; the Go runner downloads them
+// into this drive and the existing preview/fingerprint workers consume them
+// through the normal drives.Drive interface.
+package scriptcrawler
+
+import (
+	"context"
+	"errors"
+	"io"
+	"os"
+	"path/filepath"
+	"strings"
+	"time"
+
+	"github.com/video-site/backend/internal/drives"
+)
+
+const Kind = "scriptcrawler"
+
+type Config struct {
+	ID      string
+	RootDir string
+}
+
+type Driver struct {
+	id      string
+	rootDir string
+}
+
+func New(c Config) *Driver {
+	return &Driver{id: c.ID, rootDir: c.RootDir}
+}
+
+func (d *Driver) Kind() string { return Kind }
+
+func (d *Driver) ID() string { return d.id }
+
+func (d *Driver) RootID() string { return "/" }
+
+func (d *Driver) Init(context.Context) error {
+	if strings.TrimSpace(d.id) == "" {
+		return errors.New("scriptcrawler: empty drive id")
+	}
+	if strings.TrimSpace(d.rootDir) == "" {
+		return errors.New("scriptcrawler: empty root dir")
+	}
+	for _, sub := range []string{"videos", "thumbs", "output", ".crawl"} {
+		if err := os.MkdirAll(filepath.Join(d.rootDir, sub), 0o755); err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+func (d *Driver) RootDir() string { return d.rootDir }
+
+func (d *Driver) VideosDir() string { return filepath.Join(d.rootDir, "videos") }
+
+func (d *Driver) ThumbsDir() string { return filepath.Join(d.rootDir, "thumbs") }
+
+func (d *Driver) OutputDir() string { return filepath.Join(d.rootDir, "output") }
+
+func (d *Driver) CrawlDir() string { return filepath.Join(d.rootDir, ".crawl") }
+
+func (d *Driver) VideoPath(fileID string) (string, error) {
+	return safeJoin(d.VideosDir(), fileID)
+}
+
+func (d *Driver) ThumbPath(fileID string) (string, error) {
+	return safeJoin(d.ThumbsDir(), fileID)
+}
+
+func (d *Driver) OutputPath(fileName string) (string, error) {
+	return safeJoin(d.OutputDir(), fileName)
+}
+
+func (d *Driver) List(context.Context, string) ([]drives.Entry, error) {
+	entries, err := os.ReadDir(d.VideosDir())
+	if err != nil {
+		if os.IsNotExist(err) {
+			return nil, nil
+		}
+		return nil, err
+	}
+	out := make([]drives.Entry, 0, len(entries))
+	for _, e := range entries {
+		if e.IsDir() {
+			continue
+		}
+		info, err := e.Info()
+		if err != nil {
+			continue
+		}
+		out = append(out, drives.Entry{
+			ID:      e.Name(),
+			Name:    e.Name(),
+			Size:    info.Size(),
+			IsDir:   false,
+			ModTime: info.ModTime(),
+		})
+	}
+	return out, nil
+}
+
+func (d *Driver) Stat(ctx context.Context, fileID string) (*drives.Entry, error) {
+	path, err := d.VideoPath(fileID)
+	if err != nil {
+		return nil, err
+	}
+	info, err := os.Stat(path)
+	if err != nil {
+		return nil, err
+	}
+	return &drives.Entry{
+		ID:      fileID,
+		Name:    fileID,
+		Size:    info.Size(),
+		IsDir:   info.IsDir(),
+		ModTime: info.ModTime(),
+	}, nil
+}
+
+func (d *Driver) StreamURL(ctx context.Context, fileID string) (*drives.StreamLink, error) {
+	path, err := d.VideoPath(fileID)
+	if err != nil {
+		return nil, err
+	}
+	info, err := os.Stat(path)
+	if err != nil {
+		return nil, err
+	}
+	if info.IsDir() || info.Size() == 0 {
+		return nil, os.ErrNotExist
+	}
+	return &drives.StreamLink{
+		URL:     path,
+		Expires: time.Now().Add(24 * time.Hour),
+	}, nil
+}
+
+func (d *Driver) Upload(context.Context, string, string, io.Reader, int64) (string, error) {
+	return "", drives.ErrNotSupported
+}
+
+func (d *Driver) EnsureDir(context.Context, string) (string, error) {
+	return "", drives.ErrNotSupported
+}
+
+func safeJoin(root, fileID string) (string, error) {
+	id := strings.TrimSpace(fileID)
+	if id == "" || filepath.Base(id) != id {
+		return "", errors.New("scriptcrawler: invalid file id")
+	}
+	if strings.TrimSpace(root) == "" {
+		return "", errors.New("scriptcrawler: empty root")
+	}
+	rootAbs, err := filepath.Abs(root)
+	if err != nil {
+		return "", err
+	}
+	pathAbs, err := filepath.Abs(filepath.Join(rootAbs, id))
+	if err != nil {
+		return "", err
+	}
+	if pathAbs != rootAbs && !strings.HasPrefix(pathAbs, rootAbs+string(os.PathSeparator)) {
+		return "", errors.New("scriptcrawler: file id escapes root")
+	}
+	return pathAbs, nil
+}
+
+var _ drives.Drive = (*Driver)(nil)
@@ -34,6 +34,7 @@ import (
 	"github.com/video-site/backend/internal/drives/p115"
 	"github.com/video-site/backend/internal/drives/p123"
 	"github.com/video-site/backend/internal/drives/pikpak"
+	"github.com/video-site/backend/internal/drives/scriptcrawler"
 	"github.com/video-site/backend/internal/drives/spider91"
 	"github.com/video-site/backend/internal/mediaasset"
 )
@@ -58,6 +59,17 @@ type uploadTarget interface {
 	Rename(ctx context.Context, fileID, newName string) error
 }

+// Spider91LocalSource is the local source interface used by the migration
+// worker. Legacy spider91.Driver and the new scriptcrawler.Driver both satisfy
+// it when they are mounted for the Spider91 built-in crawler.
+type Spider91LocalSource interface {
+	drives.Drive
+	VideosDir() string
+	ThumbsDir() string
+	VideoPath(fileID string) (string, error)
+	ThumbPath(fileID string) (string, error)
+}
+
 // UploadResult 是 uploadTarget.UploadAndReportHash 的归一返回。
 //
 // FileID  目标盘上的新文件 ID；
@@ -364,7 +376,7 @@ func (m *Migrator) runOnce(ctx context.Context) {
 	}

 	migrated := 0
-	for _, src := range m.spider91Drives() {
+	for _, src := range m.spider91Drives(ctx) {
 		if err := ctx.Err(); err != nil {
 			return
 		}
@@ -387,7 +399,7 @@ func (m *Migrator) runOnce(ctx context.Context) {
 	// 收尾：扫每个 spider91 drive 的本地目录，把 catalog 已经迁到别处但本地
 	// 仍有残留的孤儿文件清掉。这是纯防御性兜底——正常路径下 migrateDrive
 	// 已经在迁移成功后立刻 CleanupSpider91Local，不会留孤儿。
-	for _, src := range m.spider91Drives() {
+	for _, src := range m.spider91Drives(ctx) {
 		if err := ctx.Err(); err != nil {
 			return
 		}
@@ -448,21 +460,41 @@ func (m *Migrator) resolveTarget() (string, uploadTarget, error) {
 	return id, t, nil
 }

-// spider91Drives 返回当前注册的所有 spider91 driver。
-func (m *Migrator) spider91Drives() []*spider91.Driver {
+// spider91Drives 返回当前注册的所有 Spider91 来源本地爬虫 driver。
+func (m *Migrator) spider91Drives(ctx context.Context) []Spider91LocalSource {
 	all := m.cfg.Registry.All()
-	out := make([]*spider91.Driver, 0, len(all))
+	out := make([]Spider91LocalSource, 0, len(all))
 	for _, d := range all {
-		if d.Kind() != spider91.Kind {
+		if !m.isSpider91SourceDrive(ctx, d) {
 			continue
 		}
-		if sd, ok := d.(*spider91.Driver); ok {
+		if sd, ok := d.(Spider91LocalSource); ok {
 			out = append(out, sd)
 		}
 	}
 	return out
 }

+func (m *Migrator) isSpider91SourceDrive(ctx context.Context, d drives.Drive) bool {
+	if d == nil {
+		return false
+	}
+	if d.Kind() == spider91.Kind {
+		return true
+	}
+	if d.Kind() != scriptcrawler.Kind || m.cfg.Catalog == nil {
+		return false
+	}
+	row, err := m.cfg.Catalog.GetDrive(ctx, d.ID())
+	if err != nil || row == nil {
+		return false
+	}
+	if row.Kind == spider91.Kind {
+		return true
+	}
+	return row.Kind == scriptcrawler.Kind && strings.EqualFold(strings.TrimSpace(row.Credentials["builtin"]), spider91.Kind)
+}
+
 // migrateDrive 对单个 spider91 drive 跑一批迁移；返回成功迁移的条数。
 //
 // 策略（与"本地缓存最新 N 个"语义一致）：
@@ -473,7 +505,7 @@ func (m *Migrator) spider91Drives() []*spider91.Driver {
 //   - 已经迁移过但本地还有残留 → 仅删本地（兜底）
 //
 // KeepLatestN < 0 时不保护任何本地文件，全部尝试迁移（旧行为，主要给测试用）。
-func (m *Migrator) migrateDrive(ctx context.Context, src *spider91.Driver, targetDriveID string, pp uploadTarget) (int, error) {
+func (m *Migrator) migrateDrive(ctx context.Context, src Spider91LocalSource, targetDriveID string, pp uploadTarget) (int, error) {
 	keepN := m.cfg.KeepLatestN
 	if keepN < 0 {
 		keepN = 0
@@ -574,7 +606,7 @@ func (m *Migrator) migrateDrive(ctx context.Context, src *spider91.Driver, targe
 // migrateOne 把单条 spider91 视频上传到目标盘并改写 catalog。
 // 返回 (true, nil) 表示真的迁了一条；(false, nil) 表示跳过（本地文件已不在等）；
 // (false, err) 表示真出错。
-func (m *Migrator) migrateOne(ctx context.Context, v *catalog.Video, src *spider91.Driver, targetDriveID string, pp uploadTarget) (bool, error) {
+func (m *Migrator) migrateOne(ctx context.Context, v *catalog.Video, src Spider91LocalSource, targetDriveID string, pp uploadTarget) (bool, error) {
 	path, err := src.VideoPath(v.FileID)
 	if err != nil {
 		return false, fmt.Errorf("resolve local path: %w", err)
@@ -637,7 +669,7 @@ func (m *Migrator) migrateOne(ctx context.Context, v *catalog.Video, src *spider
 	return true, nil
 }

-func (m *Migrator) preserveCrawledThumbnail(ctx context.Context, src *spider91.Driver, v *catalog.Video) {
+func (m *Migrator) preserveCrawledThumbnail(ctx context.Context, src Spider91LocalSource, v *catalog.Video) {
 	if m == nil || m.cfg.Catalog == nil || src == nil || v == nil || v.ID == "" || v.FileID == "" {
 		return
 	}
@@ -676,7 +708,7 @@ func (m *Migrator) preserveCrawledThumbnail(ctx context.Context, src *spider91.D
 	v.ThumbnailURL = "/p/thumb/" + v.ID
 }

-func findSpider91ThumbPath(src *spider91.Driver, fileID string) (string, bool) {
+func findSpider91ThumbPath(src Spider91LocalSource, fileID string) (string, bool) {
 	thumbBase := stripExt(fileID)
 	for _, ext := range []string{".jpg", ".jpeg", ".png", ".webp"} {
 		thumbPath, err := src.ThumbPath(thumbBase + ext)
@@ -722,7 +754,7 @@ func copyFileAtomic(src, dst string) error {
 // 我们不知道具体是 .jpg 还是别的，逐个尝试常见后缀）。
 //
 // 暴露成包级函数方便 cleanup 模块复用（任务 6）。
-func CleanupSpider91Local(src *spider91.Driver, fileID string) {
+func CleanupSpider91Local(src Spider91LocalSource, fileID string) {
 	videoPath, err := src.VideoPath(fileID)
 	if err == nil {
 		if err := os.Remove(videoPath); err != nil && !os.IsNotExist(err) {
@@ -759,7 +791,7 @@ func stripExt(name string) string {
 // 找到孤儿。
 //
 // 返回实际删除的文件个数。
-func (m *Migrator) cleanupOldLocalVideos(ctx context.Context, src *spider91.Driver) (int, error) {
+func (m *Migrator) cleanupOldLocalVideos(ctx context.Context, src Spider91LocalSource) (int, error) {
 	entries, err := os.ReadDir(src.VideosDir())
 	if err != nil {
 		if os.IsNotExist(err) {
@@ -17,6 +17,7 @@ import (
 	"github.com/video-site/backend/internal/drives/googledrive"
 	"github.com/video-site/backend/internal/drives/p123"
 	"github.com/video-site/backend/internal/drives/pikpak"
+	"github.com/video-site/backend/internal/drives/scriptcrawler"
 	"github.com/video-site/backend/internal/drives/spider91"
 )

@@ -599,6 +600,88 @@ func TestCleanupRemovesAllAlreadyMigratedOrphans(t *testing.T) {
 	}
 }

+func TestRunOnceMigratesBuiltInSpider91ScriptCrawlerSource(t *testing.T) {
+	ctx := context.Background()
+	cat := setupCatalog(t)
+	src := scriptcrawler.New(scriptcrawler.Config{ID: "spider-script", RootDir: t.TempDir()})
+	if err := src.Init(ctx); err != nil {
+		t.Fatalf("scriptcrawler init: %v", err)
+	}
+	if err := cat.UpsertDrive(ctx, &catalog.Drive{
+		ID:          src.ID(),
+		Kind:        scriptcrawler.Kind,
+		Name:        "Built-in Spider91",
+		Credentials: map[string]string{"builtin": "spider91"},
+	}); err != nil {
+		t.Fatalf("upsert source drive: %v", err)
+	}
+	pp := newFakePikPak("pikpak-target", "pikpak-root-id")
+	reg := newFakeRegistry()
+	reg.Add(src)
+	reg.Add(pp)
+
+	fileID := "vk-script.mp4"
+	videoPath, err := src.VideoPath(fileID)
+	if err != nil {
+		t.Fatalf("video path: %v", err)
+	}
+	if err := os.WriteFile(videoPath, []byte("scriptcrawler spider91 video"), 0o644); err != nil {
+		t.Fatalf("write video: %v", err)
+	}
+	thumbPath, err := src.ThumbPath("vk-script.jpg")
+	if err != nil {
+		t.Fatalf("thumb path: %v", err)
+	}
+	if err := os.WriteFile(thumbPath, []byte("thumb"), 0o644); err != nil {
+		t.Fatalf("write thumb: %v", err)
+	}
+	now := time.Now()
+	id := "spider91-" + src.ID() + "-vk-script"
+	if err := cat.UpsertVideo(ctx, &catalog.Video{
+		ID:            id,
+		DriveID:       src.ID(),
+		FileID:        fileID,
+		FileName:      fileID,
+		Title:         "Scriptcrawler Spider91",
+		Author:        "91porn",
+		Ext:           "mp4",
+		Quality:       "HD",
+		Size:          int64(len("scriptcrawler spider91 video")),
+		PreviewStatus: "pending",
+		PublishedAt:   now,
+		CreatedAt:     now,
+		UpdatedAt:     now,
+	}); err != nil {
+		t.Fatalf("upsert video: %v", err)
+	}
+
+	m := New(Config{
+		Catalog:          cat,
+		Registry:         reg,
+		GetTargetDriveID: func() string { return pp.ID() },
+		KeepLatestN:      -1,
+		CommonThumbDir:   t.TempDir(),
+	})
+	m.runOnce(ctx)
+
+	if pp.uploadCalls != 1 {
+		t.Fatalf("upload calls = %d, want 1", pp.uploadCalls)
+	}
+	got, err := cat.GetVideo(ctx, id)
+	if err != nil {
+		t.Fatalf("get migrated video: %v", err)
+	}
+	if got.DriveID != pp.ID() {
+		t.Fatalf("drive_id = %q, want %q", got.DriveID, pp.ID())
+	}
+	if _, err := os.Stat(videoPath); !os.IsNotExist(err) {
+		t.Fatalf("local video stat err = %v, want not exist", err)
+	}
+	if _, err := os.Stat(thumbPath); !os.IsNotExist(err) {
+		t.Fatalf("local thumb stat err = %v, want not exist", err)
+	}
+}
+
 // TestRunOnceKeepsAllLocalWhenWithinKeepWindow 验证：本地文件数 ≤ KeepLatestN 时
 // 一律不上传，全部留作"最新 N"缓存。这是用户的核心需求：刚爬下来的 15 个不要立即被传走。
 func TestRunOnceKeepsAllLocalWhenWithinKeepWindow(t *testing.T) {
@@ -0,0 +1,113 @@
+# Crawler Script Protocol v1
+
+Crawler scripts are external processes. The Go backend is the host: it handles
+dedupe, downloading, catalog writes, thumbnails, preview videos, fingerprints,
+task status and cancellation.
+
+## Invocation
+
+The backend runs:
+
+```bash
+python3 /path/to/crawler.py --job /path/to/job.json
+```
+
+`job.json`:
+
+```json
+{
+  "protocol": "crawler.v1",
+  "mode": "crawl",
+  "run_id": "20260609T120000Z",
+  "crawler_id": "example",
+  "target_new": 10,
+  "seen_source_ids_file": "/data/scriptcrawlers/example/.crawl/seen.txt",
+  "output_dir": "/data/scriptcrawlers/example/output",
+  "config": {
+    "category": "hot"
+  },
+  "network": {
+    "proxy_url": "http://127.0.0.1:7890"
+  }
+}
+```
+
+## Importing Scripts
+
+Crawler scripts are configured from the admin crawler page. A script can be
+entered as an existing server path, uploaded as a local file, or imported from
+an HTTP(S) URL.
+
+Imported scripts are copied into `crawler-scripts/` next to the configured local
+preview data directory. The import API currently accepts Python files only
+(`.py`) and rejects empty files or files larger than 2 MiB.
+
+## Output
+
+stdout must be JSON Lines. Logs must go to stderr.
+
+Recommended item event:
+
+```json
+{
+  "type": "item",
+  "title": "Video title",
+  "media_url": "https://cdn.example.test/video.mp4",
+  "thumbnail_url": "https://cdn.example.test/cover.jpg",
+  "source_id": "site-native-id",
+  "headers": {
+    "Referer": "https://example.test/"
+  }
+}
+```
+
+Minimum item event:
+
+```json
+{"type":"item","title":"Video title","media_url":"https://cdn.example.test/video.mp4"}
+```
+
+If a line contains item fields such as `title` and `media_url`, the backend also
+treats it as an item when `type` is omitted.
+
+The item fields may also be wrapped inside `"item"` if that is more convenient:
+
+```json
+{"type":"item","item":{"title":"Video title","media_url":"https://cdn.example.test/video.mp4"}}
+```
+
+Optional progress/done events:
+
+```json
+{"type":"progress","checked":20,"emitted":3}
+{"type":"done","stats":{"emitted":10}}
+```
+
+## Simple Field Rules
+
+- `title` is required.
+- `media_url` is required for normal scripts. The backend downloads the video.
+- `thumbnail_url` is optional. If it is empty, the backend generates a thumbnail
+  from the downloaded video.
+- `source_id` is optional but recommended. If present, it should be stable
+  within one crawler and lets the backend skip known videos before downloading.
+  If it is empty, the backend creates an internal `auto-...` ID and later relies
+  on the existing video fingerprint dedupe path.
+- `headers` is optional and is applied to both video and thumbnail downloads.
+  Use it for `Referer`, cookies or anti-hotlinking requirements.
+
+## Advanced Fields
+
+- `detail_url`, `author`, `tags`, `category`, `quality`, `duration_seconds`,
+  `description` and `published_at` are optional metadata fields.
+- If video and thumbnail need different headers, use `media_headers` and
+  `thumbnail_headers`.
+- Existing nested fields are still supported for compatibility:
+  `media.url`, `media.local_file`, `media.headers`, `thumbnail.url`,
+  `thumbnail.local_file`, `thumbnail.headers`.
+- Advanced scripts may download into `job.output_dir` and return
+  `media_local_file` or `media.local_file`. The path must stay inside
+  `output_dir`.
+- Scripts can read `seen_source_ids_file` and skip known IDs when they provide
+  stable `source_id` values. The backend still dedupes every item.
+- The backend stops the process after `target_new` new videos are imported.
@@ -0,0 +1,48 @@
+#!/usr/bin/env python3
+import argparse
+import json
+import sys
+
+
+def load_seen(path):
+    try:
+        with open(path, "r", encoding="utf-8") as f:
+            return {line.strip() for line in f if line.strip()}
+    except FileNotFoundError:
+        return set()
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--job", required=True)
+    args = parser.parse_args()
+
+    with open(args.job, "r", encoding="utf-8") as f:
+        job = json.load(f)
+
+    seen = load_seen(job.get("seen_source_ids_file", ""))
+    source_id = "demo-video-1"
+    if source_id in seen:
+        print(json.dumps({"type": "done", "stats": {"emitted": 0}}), flush=True)
+        return
+
+    event = {
+        "type": "item",
+        "source_id": source_id,
+        "title": "Demo Video",
+        "media_url": "https://example.test/video/demo-video-1.mp4",
+        "thumbnail_url": "https://example.test/thumb/demo-video-1.jpg",
+        "headers": {
+            "Referer": "https://example.test/",
+        },
+    }
+    print(json.dumps(event, ensure_ascii=False), flush=True)
+    print(json.dumps({"type": "done", "stats": {"emitted": 1}}), flush=True)
+
+
+if __name__ == "__main__":
+    try:
+        main()
+    except Exception as exc:
+        print(f"crawler failed: {exc}", file=sys.stderr, flush=True)
+        raise
@@ -8,6 +8,7 @@ import { AdminLayout } from "@/admin/AdminLayout";
 import { LoginPage } from "@/admin/LoginPage";
 import { RequireAuth } from "@/admin/RequireAuth";
 import { DrivesPage } from "@/admin/DrivesPage";
+import { CrawlersPage } from "@/admin/CrawlersPage";
 import { VideosPage } from "@/admin/VideosPage";
 import { TagsPage } from "@/admin/TagsPage";
 import { ThemePage } from "@/admin/ThemePage";
@@ -70,6 +71,7 @@ export default function App() {
      >
        <Route index element={<Navigate to="/admin/drives" replace />} />
        <Route path="drives" element={<DrivesPage />} />
+        <Route path="crawlers" element={<CrawlersPage />} />
        <Route path="videos" element={<VideosPage />} />
        <Route path="tags" element={<TagsPage />} />
        <Route path="theme" element={<ThemePage />} />
@@ -14,6 +14,7 @@ import {
 import * as api from "./api";
 import { useAuth } from "./AuthContext";
 import { useToast } from "./ToastContext";
+import { SpiderIcon } from "./icons/SpiderIcon";

 export function AdminLayout() {
  const { logout } = useAuth();
@@ -88,6 +89,14 @@ export function AdminLayout() {
          >
            <HardDrive size={16} /> 网盘管理
          </NavLink>
+          <NavLink
+            to="/admin/crawlers"
+            className={({ isActive }) =>
+              `admin-nav__link ${isActive ? "is-active" : ""}`
+            }
+          >
+            <SpiderIcon size={16} /> 爬虫管理
+          </NavLink>
          <NavLink
            to="/admin/videos"
            className={({ isActive }) =>
@@ -0,0 +1,402 @@
+import { useEffect, useMemo, useState } from "react";
+import { ArrowLeft, CircleStop, Download, Link as LinkIcon, Plus, Save, Trash2, Upload } from "lucide-react";
+import * as api from "./api";
+import { useToast } from "./ToastContext";
+import { driveKindAbbr, generationStateClass, generationStateLabel } from "./drive/constants";
+import { SpiderIcon } from "./icons/SpiderIcon";
+
+type CrawlerForm = {
+  id: string;
+  name: string;
+  builtin: string;
+  scriptPath: string;
+  pythonPath: string;
+  targetNew: string;
+  proxy: string;
+  configJson: string;
+};
+
+const emptyForm: CrawlerForm = {
+  id: "",
+  name: "",
+  builtin: "",
+  scriptPath: "",
+  pythonPath: "python3",
+  targetNew: "10",
+  proxy: "",
+  configJson: "",
+};
+
+export function CrawlersPage() {
+  const [list, setList] = useState<api.AdminCrawler[]>([]);
+  const [selectedId, setSelectedId] = useState("");
+  const [form, setForm] = useState<CrawlerForm>(emptyForm);
+  const [loading, setLoading] = useState(true);
+  const [saving, setSaving] = useState(false);
+  const [runningId, setRunningId] = useState("");
+  const [stoppingId, setStoppingId] = useState("");
+  const [scriptURL, setScriptURL] = useState("");
+  const [importingScript, setImportingScript] = useState(false);
+  const [mode, setMode] = useState<"list" | "detail">("list");
+  const { show } = useToast();
+
+  const selected = useMemo(
+    () => list.find((item) => item.id === selectedId) ?? null,
+    [list, selectedId]
+  );
+
+  async function refresh() {
+    setLoading(true);
+    try {
+      const data = await api.listCrawlers();
+      setList(data);
+    } catch (e) {
+      show(e instanceof Error ? e.message : "加载爬虫失败", "error");
+    } finally {
+      setLoading(false);
+    }
+  }
+
+  useEffect(() => {
+    refresh();
+  }, []);
+
+  function selectCrawler(crawler: api.AdminCrawler) {
+    setSelectedId(crawler.id);
+    setMode("detail");
+    setForm({
+      id: crawler.id,
+      name: crawler.name,
+      builtin: crawler.builtin ?? "",
+      scriptPath: crawler.scriptPath ?? "",
+      pythonPath: crawler.pythonPath || "python3",
+      targetNew: crawler.targetNew || (crawler.builtin === "spider91" || crawler.kind === "spider91" ? "15" : "10"),
+      proxy: crawler.proxy ?? "",
+      configJson: crawler.configJson ?? "",
+    });
+  }
+
+  function createCustom() {
+    setSelectedId("");
+    setForm(emptyForm);
+    setScriptURL("");
+    setMode("detail");
+  }
+
+  function createSpider91() {
+    setSelectedId("");
+    setForm({
+      ...emptyForm,
+      id: "spider91",
+      name: "91 爬虫",
+      builtin: "spider91",
+      scriptPath: "",
+      targetNew: "15",
+    });
+    setScriptURL("");
+    setMode("detail");
+  }
+
+  function backToList() {
+    setSelectedId("");
+    setForm(emptyForm);
+    setScriptURL("");
+    setMode("list");
+  }
+
+  function set<K extends keyof CrawlerForm>(key: K, value: CrawlerForm[K]) {
+    setForm((prev) => ({ ...prev, [key]: value }));
+  }
+
+  async function save() {
+    const id = form.id.trim();
+    const name = form.name.trim();
+    if (!id || !name) {
+      show("请填写爬虫 ID 和名称", "error");
+      return;
+    }
+    if (!form.builtin && !form.scriptPath.trim()) {
+      show("请先导入爬虫脚本", "error");
+      return;
+    }
+    setSaving(true);
+    try {
+      const resp = await api.upsertCrawler({
+        id,
+        name,
+        builtin: form.builtin,
+        scriptPath: form.scriptPath.trim(),
+        pythonPath: form.pythonPath.trim(),
+        targetNew: form.targetNew.trim(),
+        proxy: form.proxy.trim(),
+        configJson: form.configJson.trim(),
+      });
+      if (resp.warning) {
+        show(`已保存，但初始化失败：${resp.warning}`, "error");
+      } else {
+        show("已保存", "success");
+      }
+      setSelectedId(id);
+      await refresh();
+      setMode("list");
+    } catch (e) {
+      show(e instanceof Error ? e.message : "保存失败", "error");
+    } finally {
+      setSaving(false);
+    }
+  }
+
+  async function importScriptFile(file: File | null | undefined) {
+    if (!file) return;
+    setImportingScript(true);
+    try {
+      const resp = await api.importCrawlerScriptFile(file);
+      set("scriptPath", resp.scriptPath);
+      show("脚本已导入", "success");
+    } catch (e) {
+      show(e instanceof Error ? e.message : "导入失败", "error");
+    } finally {
+      setImportingScript(false);
+    }
+  }
+
+  async function importScriptURL() {
+    const url = scriptURL.trim();
+    if (!url) {
+      show("请填写脚本链接", "error");
+      return;
+    }
+    setImportingScript(true);
+    try {
+      const resp = await api.importCrawlerScriptURL(url);
+      set("scriptPath", resp.scriptPath);
+      setScriptURL("");
+      show("脚本已导入", "success");
+    } catch (e) {
+      show(e instanceof Error ? e.message : "导入失败", "error");
+    } finally {
+      setImportingScript(false);
+    }
+  }
+
+  async function run(crawler: api.AdminCrawler) {
+    setRunningId(crawler.id);
+    try {
+      const resp = await api.runCrawler(crawler.id);
+      if (!resp.accepted) {
+        show(resp.message || "当前爬虫有正在进行的任务", "info");
+        return;
+      }
+      show("已触发抓取任务", "success");
+      await refresh();
+    } catch (e) {
+      show(e instanceof Error ? e.message : "触发失败", "error");
+    } finally {
+      setRunningId("");
+    }
+  }
+
+  async function stop(crawler: api.AdminCrawler) {
+    setStoppingId(crawler.id);
+    try {
+      const resp = await api.stopCrawlerTasks(crawler.id);
+      show(resp.stopped ? "已请求停止任务" : "当前没有可停止任务", "info");
+      await refresh();
+    } catch (e) {
+      show(e instanceof Error ? e.message : "停止失败", "error");
+    } finally {
+      setStoppingId("");
+    }
+  }
+
+  async function remove(crawler: api.AdminCrawler) {
+    if (!window.confirm(`删除爬虫 ${crawler.name} 并清理它导入的视频？`)) return;
+    try {
+      const resp = await api.deleteCrawler(crawler.id);
+      show(`已删除，并清理 ${resp.deletedVideos ?? 0} 个视频`, "success");
+      setSelectedId("");
+      setForm(emptyForm);
+      setMode("list");
+      await refresh();
+    } catch (e) {
+      show(e instanceof Error ? e.message : "删除失败", "error");
+    }
+  }
+
+  return (
+    <section className="admin-page">
+      <header className="admin-page__header">
+        <div>
+          <h1 className="admin-page__title">爬虫管理</h1>
+        </div>
+        <div className="admin-detail-actions-inline">
+          {mode === "list" ? (
+            <button className="admin-btn is-primary" onClick={createCustom}>
+              <Plus size={14} /> 添加爬虫
+            </button>
+          ) : (
+            <button className="admin-btn" onClick={backToList}>
+              <ArrowLeft size={14} /> 返回列表
+            </button>
+          )}
+        </div>
+      </header>
+
+      {mode === "list" ? (
+        <div className="admin-card admin-crawler-list">
+          <header className="admin-card__title">
+            <SpiderIcon size={16} /> 已配置爬虫
+          </header>
+          {loading ? (
+            <div className="admin-loading">加载中...</div>
+          ) : list.length === 0 ? (
+            <div className="admin-empty">暂无爬虫</div>
+          ) : (
+            <div className="admin-drive-teasers">
+              {list.map((crawler) => (
+                <button
+                  key={crawler.id}
+                  type="button"
+                  className={`admin-drive-teaser ${crawler.id === selectedId ? "is-active" : ""}`}
+                  onClick={() => selectCrawler(crawler)}
+                >
+                  <span className="admin-drive-teaser__name">
+                    <span className="admin-drive-card__brand-icon" data-kind={crawler.builtin || crawler.kind}>
+                      {crawler.builtin === "spider91" ? "91" : driveKindAbbr(crawler.kind)}
+                    </span>
+                    {crawler.name}
+                  </span>
+                  <span className={`admin-status is-${crawler.status === "ok" ? "ok" : crawler.status === "error" ? "error" : "pending"}`}>
+                    {crawler.status === "ok" ? "已就绪" : crawler.status === "error" ? "错误" : "未连接"}
+                  </span>
+                </button>
+              ))}
+            </div>
+          )}
+        </div>
+      ) : (
+        <div className="admin-crawler-detail">
+          <div className="admin-card">
+            <header className="admin-card__title">
+              <SpiderIcon size={16} /> {selected ? "爬虫配置" : "添加爬虫"}
+            </header>
+            <div className="admin-form">
+              {!selected && (
+                <div className="admin-crawler-presets">
+                  <button className={`admin-btn ${form.builtin === "" ? "is-primary" : ""}`} type="button" onClick={createCustom}>
+                    <Plus size={13} /> 自定义脚本
+                  </button>
+                  <button className={`admin-btn ${form.builtin === "spider91" ? "is-primary" : ""}`} type="button" onClick={createSpider91}>
+                    <SpiderIcon size={13} /> 内置 91
+                  </button>
+                </div>
+              )}
+              <div className="admin-form__row">
+                <label htmlFor="crawler-id">爬虫 ID *</label>
+                <input id="crawler-id" value={form.id} onChange={(e) => set("id", e.target.value)} disabled={!!selected} />
+              </div>
+              <div className="admin-form__row">
+                <label htmlFor="crawler-name">名称 *</label>
+                <input id="crawler-name" value={form.name} onChange={(e) => set("name", e.target.value)} />
+              </div>
+              {!form.builtin && (
+                <div className="admin-form__row">
+                  <label htmlFor="crawler-script-url">导入脚本</label>
+                  <div className="admin-crawler-import">
+                    <input
+                      id="crawler-script-file"
+                      className="admin-crawler-import__file"
+                      type="file"
+                      accept=".py,text/x-python"
+                      disabled={importingScript}
+                      onChange={(e) => {
+                        importScriptFile(e.target.files?.[0]);
+                        e.currentTarget.value = "";
+                      }}
+                    />
+                    <label className="admin-btn" htmlFor="crawler-script-file" aria-disabled={importingScript}>
+                      <Upload size={13} /> 上传文件
+                    </label>
+                    <input
+                      id="crawler-script-url"
+                      value={scriptURL}
+                      onChange={(e) => setScriptURL(e.target.value)}
+                      placeholder="https://example.com/crawler.py"
+                      disabled={importingScript}
+                    />
+                    <button className="admin-btn" type="button" onClick={importScriptURL} disabled={importingScript}>
+                      <LinkIcon size={13} /> {importingScript ? "导入中..." : "链接导入"}
+                    </button>
+                  </div>
+                  {form.scriptPath && <div className="admin-form__help">脚本已导入</div>}
+                </div>
+              )}
+              <div className="admin-form__row">
+                <label htmlFor="crawler-target">每次补充新视频数</label>
+                <input id="crawler-target" value={form.targetNew} onChange={(e) => set("targetNew", e.target.value)} placeholder="10" />
+              </div>
+              <div className="admin-form__row">
+                <label htmlFor="crawler-proxy">代理地址</label>
+                <input id="crawler-proxy" value={form.proxy} onChange={(e) => set("proxy", e.target.value)} placeholder="http://127.0.0.1:7890" />
+              </div>
+              <div className="admin-detail-actions">
+                <button className="admin-btn is-primary" onClick={save} disabled={saving}>
+                  <Save size={13} /> {saving ? "保存中..." : "保存"}
+                </button>
+                {selected && (
+                  <>
+                    <button className="admin-btn" onClick={() => run(selected)} disabled={runningId === selected.id}>
+                      <Download size={13} /> {runningId === selected.id ? "触发中..." : "立即抓取"}
+                    </button>
+                    <button className="admin-btn is-stop" onClick={() => stop(selected)} disabled={stoppingId === selected.id}>
+                      <CircleStop size={13} /> {stoppingId === selected.id ? "停止中..." : "停止任务"}
+                    </button>
+                    <button className="admin-btn is-danger" onClick={() => remove(selected)}>
+                      <Trash2 size={13} /> 删除
+                    </button>
+                  </>
+                )}
+              </div>
+            </div>
+          </div>
+
+          {selected && (
+            <div className="admin-card admin-crawler-status">
+              <header className="admin-card__title">
+                <Download size={16} /> 状态
+              </header>
+              <div className="admin-gen-columns">
+                <CrawlerStatus label="抓取" status={selected.scanGenerationStatus} />
+                <CrawlerStatus label="封面" status={selected.thumbnailGenerationStatus} />
+                <CrawlerStatus label="预览视频" status={selected.previewGenerationStatus} />
+                <CrawlerStatus label="视频指纹" status={selected.fingerprintGenerationStatus} />
+              </div>
+              {selected.lastError && <div className="admin-detail-error">{selected.lastError}</div>}
+            </div>
+          )}
+        </div>
+      )}
+    </section>
+  );
+}
+
+function CrawlerStatus({ label, status }: { label: string; status?: api.DriveGenerationStatus }) {
+  const state = status?.state || "idle";
+  const labelText = label === "抓取" && state === "scanning" ? "抓取中" : generationStateLabel(state);
+  return (
+    <div className="admin-gen-col">
+      <div className="admin-gen-col__head">
+        <span className="admin-gen-col__label">{label}</span>
+        <span className={`admin-status admin-generation-state is-${generationStateClass(state)}`}>
+          {labelText}
+        </span>
+      </div>
+      {label === "抓取" && (
+        <div className="admin-gen-col__counts admin-gen-col__counts--scan">
+          <div className="admin-gen-col__count"><span>已抓取</span><strong>{status?.scannedCount ?? 0}</strong></div>
+          <div className="admin-gen-col__count"><span>预计新增</span><strong>{status?.addedCount ?? 0}</strong></div>
+        </div>
+      )}
+    </div>
+  );
+}
@@ -325,6 +325,10 @@ export function DrivesPage() {
  }

  async function handleRescan(d: api.AdminDrive) {
+    if (d.kind === "spider91") {
+      show("91Spider 不再支持通过网盘运行，请到爬虫管理添加爬虫脚本", "info");
+      return;
+    }
    if (nightlyBusy) {
      show(nightlyBusyText(nightlyStatus) || NIGHTLY_BUSY_MESSAGE, "info");
      return;
@@ -345,11 +349,7 @@ export function DrivesPage() {
        refreshDriveList();
        return;
      }
-      if (d.kind === "spider91") {
-        show("已触发抓取任务，需要 2-4 分钟，可稍后刷新视频列表查看", "success");
-      } else {
-        show("已触发扫描，可稍后刷新视频列表查看", "success");
-      }
+      show("已触发扫描，可稍后刷新视频列表查看", "success");
      refreshDriveList();
    } catch (e) {
      show(e instanceof Error ? e.message : "触发失败", "error");
@@ -550,10 +550,8 @@ export function DrivesPage() {
                )}
                {d.kind === "spider91" && (
                  <div className="admin-detail-row">
-                    <span className="admin-detail-label">上次抓取时间</span>
-                    <span className="admin-detail-value">
-                      {d.lastCrawlAt ? new Date(d.lastCrawlAt * 1000).toLocaleString() : "尚未抓取"}
-                    </span>
+                    <span className="admin-detail-label">配置状态</span>
+                    <span className="admin-detail-value">已废弃，请到爬虫管理添加</span>
                  </div>
                )}
              </div>
@@ -567,9 +565,12 @@ export function DrivesPage() {
                    type="button"
                    className="admin-btn is-primary"
                    onClick={() => handleRescan(d)}
-                    aria-disabled={nightlyBusy || isDriveBusy(d) || !!scanningDriveIds[d.id]}
+                    disabled={d.kind === "spider91"}
+                    aria-disabled={d.kind === "spider91" || nightlyBusy || isDriveBusy(d) || !!scanningDriveIds[d.id]}
                    title={
-                      nightlyBusy
+                      d.kind === "spider91"
+                        ? "91Spider 不再支持通过网盘运行，请到爬虫管理添加爬虫脚本"
+                        : nightlyBusy
                        ? nightlyBusyText(nightlyStatus) || NIGHTLY_BUSY_MESSAGE
                        : isDriveBusy(d) || scanningDriveIds[d.id]
                        ? DRIVE_BUSY_MESSAGE
@@ -579,7 +580,7 @@ export function DrivesPage() {
                    {d.kind === "spider91" ? (
                      <>
                        <Download size={13} className={scanningDriveIds[d.id] ? "admin-spin" : undefined} />
-                        {scanningDriveIds[d.id] ? "触发中..." : "立即抓取"}
+                        已废弃
                      </>
                    ) : (
                      <>
@@ -599,9 +600,11 @@ export function DrivesPage() {
                    {stoppingDriveId === d.id ? "停止中..." : "停止所有任务"}
                  </button>
                </div>
-                <button type="button" className="admin-btn" onClick={() => openEdit(d)}>
-                  {d.kind === "spider91" ? "编辑配置" : "编辑配置凭证"}
-                </button>
+                {d.kind !== "spider91" && (
+                  <button type="button" className="admin-btn" onClick={() => openEdit(d)}>
+                    编辑配置凭证
+                  </button>
+                )}
                <button type="button" className="admin-btn is-danger admin-detail-actions__danger" onClick={() => setDeleteTarget(d)}>
                  <Trash2 size={13} /> 删除网盘
                </button>
@@ -12,13 +12,14 @@ async function request<T>(
  path: string,
  init: RequestInit = {}
 ): Promise<T> {
+  const headers = new Headers(init.headers ?? {});
+  if (!(init.body instanceof FormData) && !headers.has("Content-Type")) {
+    headers.set("Content-Type", "application/json");
+  }
  const res = await fetch(BASE + path, {
    credentials: "include",
-    headers: {
-      "Content-Type": "application/json",
-      ...(init.headers ?? {}),
-    },
    ...init,
+    headers,
  });
  if (res.status === 401) {
    throw new UnauthorizedError();
@@ -188,6 +189,99 @@ export function stopDriveTasks(id: string) {
  );
 }

+// ---------- Crawlers ----------
+
+export type AdminCrawler = {
+  id: string;
+  name: string;
+  kind: "scriptcrawler" | "spider91";
+  builtin?: string;
+  status: string;
+  lastError?: string;
+  scriptPath: string;
+  pythonPath?: string;
+  proxy?: string;
+  targetNew?: string;
+  configJson?: string;
+  lastCrawlAt?: number;
+  scanGenerationStatus?: DriveGenerationStatus;
+  thumbnailGenerationStatus?: DriveGenerationStatus;
+  previewGenerationStatus?: DriveGenerationStatus;
+  fingerprintGenerationStatus?: DriveGenerationStatus;
+  thumbnailReadyCount: number;
+  thumbnailPendingCount: number;
+  thumbnailFailedCount: number;
+  teaserReadyCount: number;
+  teaserPendingCount: number;
+  teaserFailedCount: number;
+  fingerprintReadyCount: number;
+  fingerprintPendingCount: number;
+  fingerprintFailedCount: number;
+};
+
+export type UpsertCrawlerInput = {
+  id: string;
+  name: string;
+  builtin?: string;
+  scriptPath: string;
+  pythonPath?: string;
+  proxy?: string;
+  targetNew?: string;
+  configJson?: string;
+};
+
+export type ImportCrawlerScriptResult = {
+  scriptPath: string;
+};
+
+export function listCrawlers() {
+  return request<AdminCrawler[]>("/crawlers");
+}
+
+export function upsertCrawler(body: UpsertCrawlerInput) {
+  return request<{ ok: boolean; warning?: string }>("/crawlers", {
+    method: "POST",
+    body: JSON.stringify(body),
+  });
+}
+
+export function importCrawlerScriptFile(file: File) {
+  const form = new FormData();
+  form.append("file", file);
+  return request<ImportCrawlerScriptResult>("/crawlers/import-file", {
+    method: "POST",
+    body: form,
+  });
+}
+
+export function importCrawlerScriptURL(url: string) {
+  return request<ImportCrawlerScriptResult>("/crawlers/import-url", {
+    method: "POST",
+    body: JSON.stringify({ url }),
+  });
+}
+
+export function runCrawler(id: string) {
+  return request<{ ok: boolean; accepted: boolean; message?: string; status?: NightlyJobStatus }>(
+    `/crawlers/${encodeURIComponent(id)}/run`,
+    { method: "POST" }
+  );
+}
+
+export function stopCrawlerTasks(id: string) {
+  return request<{ ok: boolean; stopped: boolean }>(
+    `/crawlers/${encodeURIComponent(id)}/tasks/stop`,
+    { method: "POST" }
+  );
+}
+
+export function deleteCrawler(id: string) {
+  return request<{ ok: boolean; deletedVideos: number }>(`/crawlers/${encodeURIComponent(id)}`, {
+    method: "DELETE",
+    body: JSON.stringify({ deleteVideos: true }),
+  });
+}
+
 export type P123QRSession = {
  loginUuid: string;
  uniID: string;
@@ -101,13 +101,17 @@ export function StatusTag({
  error?: string;
  hasCred: boolean;
 }) {
+  if (kind === "spider91") {
+    return (
+      <span className="admin-status is-error" title={error || "请到爬虫管理添加爬虫脚本"}>
+        已废弃
+      </span>
+    );
+  }
  if (kind !== "spider91" && !hasCred) {
    return <span className="admin-status is-pending">未配置凭证</span>;
  }
  if (status === "ok") {
-    if (kind === "spider91") {
-      return <span className="admin-status is-ok">已就绪</span>;
-    }
    return <span className="admin-status is-ok">已连接</span>;
  }
  if (status === "error")
@@ -205,7 +209,7 @@ export function DriveGenerationPanel({

      <div className="admin-gen-columns">
        <DriveGenCol
-          label={d.kind === "spider91" ? "抓取" : "扫盘"}
+          label={d.kind === "spider91" ? "已废弃" : "扫盘"}
          status={d.scanGenerationStatus}
          showCounts={false}
        />
@@ -26,7 +26,6 @@ const DRIVE_OPTIONS: DriveOption[] = [
  { kind: "onedrive", label: "OneDrive", abbr: "OD", desc: "302直链，微软网盘" },
  { kind: "googledrive", label: "Google Drive", abbr: "GD", desc: "服务器中转模式" },
  { kind: "localstorage", label: "本地存储", abbr: "Lo", desc: "本机文件目录" },
-  { kind: "spider91", label: "91 爬虫", abbr: "91", desc: "自动抓取热门视频" },
  { kind: "quark", label: "夸克网盘", abbr: "Qk", desc: "302直链" },
  { kind: "wopan", label: "联通沃盘", abbr: "Wo", desc: "302直链" },
 ];
@@ -163,7 +163,7 @@ export function credentialHelp(kind: Kind, isEdit: boolean): string {
    case "localstorage":
      return `填写服务器可访问的本地目录绝对路径，例如 /mnt/videos。系统会扫描该目录及子目录中的视频文件和 .strm 文件；.strm 可指向 HTTP/HTTPS 直链，或指向本地存储根目录内的真实视频路径。Docker 部署时请填写容器内路径。${note}`;
    case "spider91":
-      return "91 爬虫会把定时抓取到的视频和封面先保存到本机，并作为一个视频来源接入站点；可按服务器网络情况单独配置代理。后续流水线会把较早的视频上传到你选择的 115 / PikPak / OneDrive 目标盘。";
+      return "91Spider 不再支持通过网盘添加或编辑。请到后台爬虫管理页面添加内置 91 或自定义爬虫脚本。";
    default:
      return "";
  }
@@ -0,0 +1,34 @@
+type SpiderIconProps = {
+  size?: number;
+  className?: string;
+};
+
+export function SpiderIcon({ size = 16, className }: SpiderIconProps) {
+  return (
+    <svg
+      className={className}
+      width={size}
+      height={size}
+      viewBox="0 0 24 24"
+      fill="none"
+      stroke="currentColor"
+      strokeWidth="2"
+      strokeLinecap="round"
+      strokeLinejoin="round"
+      aria-hidden="true"
+      focusable="false"
+    >
+      <path d="M12 7v3" />
+      <path d="M9 5.5 7.5 3" />
+      <path d="M15 5.5 16.5 3" />
+      <path d="M9 10.5 4.5 8" />
+      <path d="M15 10.5 19.5 8" />
+      <path d="M8.5 13.5 3 13" />
+      <path d="M15.5 13.5 21 13" />
+      <path d="M9 16 5 20" />
+      <path d="M15 16 19 20" />
+      <ellipse cx="12" cy="14" rx="4" ry="5" />
+      <circle cx="12" cy="8" r="2.5" />
+    </svg>
+  );
+}
@@ -308,6 +308,33 @@
  margin-bottom: var(--space-3);
 }

+.admin-crawler-layout {
+  display: grid;
+  grid-template-columns: minmax(260px, 0.8fr) minmax(360px, 1.2fr);
+  gap: var(--space-4);
+  align-items: start;
+}
+
+.admin-crawler-detail {
+  display: grid;
+  gap: var(--space-4);
+}
+
+.admin-crawler-status {
+  grid-column: 1 / -1;
+}
+
+.admin-crawler-list .admin-drive-teaser {
+  width: 100%;
+}
+
+.admin-crawler-presets {
+  display: flex;
+  flex-wrap: wrap;
+  gap: var(--space-2);
+  margin-bottom: var(--space-3);
+}
+
 /* ----- Storage summary ----- */
 .admin-storage-summary {
  display: grid;
@@ -446,6 +473,21 @@
  align-items: center;
 }

+.admin-crawler-import {
+  display: grid;
+  grid-template-columns: auto minmax(180px, 1fr) auto;
+  gap: var(--space-2);
+  align-items: center;
+}
+
+.admin-crawler-import__file {
+  position: absolute;
+  width: 1px;
+  height: 1px;
+  opacity: 0;
+  pointer-events: none;
+}
+
 .admin-form__help {
  font-size: var(--font-xs);
  color: var(--text-faint);
@@ -1548,6 +1590,15 @@
    width: 100%;
  }

+  .admin-crawler-import {
+    grid-template-columns: 1fr;
+  }
+
+  .admin-crawler-import .admin-btn {
+    justify-content: center;
+    width: 100%;
+  }
+
  .admin-p123-qr__body {
    grid-template-columns: 1fr;
  }
@@ -3332,6 +3383,7 @@
 }

@media (max-width: 900px) {
+  .admin-crawler-layout,
  .admin-tags-layout {
    grid-template-columns: 1fr;
  }
@@ -10,6 +10,18 @@ const driveComponentsSource = readFileSync(
  new URL("../src/admin/drive/DriveComponents.tsx", import.meta.url),
  "utf8"
 );
+const crawlerPageSource = readFileSync(
+  new URL("../src/admin/CrawlersPage.tsx", import.meta.url),
+  "utf8"
+);
+const adminLayoutSource = readFileSync(
+  new URL("../src/admin/AdminLayout.tsx", import.meta.url),
+  "utf8"
+);
+const appSource = readFileSync(
+  new URL("../src/App.tsx", import.meta.url),
+  "utf8"
+);
 const spider91UploadTargetSource = readFileSync(
  new URL("../src/admin/drive/Spider91UploadTargetField.tsx", import.meta.url),
  "utf8"
@@ -51,14 +63,15 @@ function assertDriveTypeOption(value: string, label: string) {
  );
 }

-test("spider91 drive form does not expose advanced crawler credentials", () => {
-  assert.match(combinedSource, /key: "proxy"/);
-  assert.match(combinedSource, /label: "代理地址（可选）"/);
-  assert.match(combinedSource, /支持 http:\/\/、https:\/\/、socks5:\/\/、socks5h:\/\/代理/);
-  assert.doesNotMatch(combinedSource, /target_new/);
-  assert.doesNotMatch(combinedSource, /crawl_hour/);
-  assert.doesNotMatch(combinedSource, /python_path/);
-  assert.doesNotMatch(combinedSource, /script_path/);
+test("crawler sources are not selectable as storage drives", () => {
+  assert.ok(
+    !driveTypeOptions().some((option) => option.value === "spider91"),
+    "spider91 should not be a storage drive option"
+  );
+  assert.ok(
+    !driveTypeOptions().some((option) => option.value === "scriptcrawler"),
+    "scriptcrawler should not be a storage drive option"
+  );
 });

 test("spider91 upload target uses explicit local-save option instead of auto target", () => {
@@ -185,12 +198,45 @@ test("drive type selector keeps primary source order", () => {
    { value: "onedrive", label: "OneDrive" },
    { value: "googledrive", label: "Google Drive" },
    { value: "localstorage", label: "本地存储" },
-    { value: "spider91", label: "91 爬虫" },
    { value: "quark", label: "夸克网盘" },
    { value: "wopan", label: "联通沃盘" },
  ]);
 });

+test("crawler management is a separate admin section", () => {
+  assert.match(adminLayoutSource, /to="\/admin\/crawlers"/);
+  assert.match(adminLayoutSource, /> 爬虫管理/);
+  assert.match(adminLayoutSource, /SpiderIcon size=\{16\} \/> 爬虫管理/);
+  assert.match(appSource, /path="crawlers" element=\{<CrawlersPage \/>/);
+  assert.match(crawlerPageSource, /export function CrawlersPage/);
+  assert.match(crawlerPageSource, /SpiderIcon/);
+  assert.match(crawlerPageSource, /添加爬虫/);
+  assert.match(crawlerPageSource, /返回列表/);
+  assert.match(crawlerPageSource, /setMode\("detail"\)/);
+  assert.match(crawlerPageSource, /setMode\("list"\)/);
+  assert.match(crawlerPageSource, /api\.listCrawlers/);
+  assert.match(crawlerPageSource, /api\.upsertCrawler/);
+  assert.match(crawlerPageSource, /api\.runCrawler/);
+  assert.match(crawlerPageSource, /api\.stopCrawlerTasks/);
+  assert.match(crawlerPageSource, /api\.deleteCrawler/);
+  assert.match(crawlerPageSource, /api\.importCrawlerScriptFile/);
+  assert.match(crawlerPageSource, /api\.importCrawlerScriptURL/);
+  assert.match(crawlerPageSource, /type="file"/);
+  assert.match(crawlerPageSource, /链接导入/);
+  assert.doesNotMatch(crawlerPageSource, /新建脚本/);
+  assert.doesNotMatch(crawlerPageSource, /脚本路径/);
+  assert.doesNotMatch(crawlerPageSource, /Python 解释器/);
+  assert.doesNotMatch(crawlerPageSource, /自定义配置 JSON/);
+  assert.doesNotMatch(crawlerPageSource, /Bot/);
+  assert.match(crawlerPageSource, /builtin:\s*"spider91"/);
+  assert.match(apiSource, /type AdminCrawler/);
+  assert.match(apiSource, /"\/crawlers"/);
+  assert.match(apiSource, /"\/crawlers\/import-file"/);
+  assert.match(apiSource, /"\/crawlers\/import-url"/);
+  assert.match(apiSource, /new FormData\(\)/);
+  assert.doesNotMatch(driveFormSource, /scriptcrawler/);
+});
+
 test("drive cards use configured abbreviations and visible fallback icon colors", () => {
  assert.match(constantsSource, /googledrive:\s*"GD"/);
  assert.match(constantsSource, /function driveKindAbbr\(kind: string\)/);
@@ -230,10 +276,9 @@ test("nightly scan duplicate trigger uses full-scan busy message", () => {
 });

 test("drive generation panel shows scan or crawler status first", () => {
-  assert.match(driveComponentsSource, /label=\{d\.kind === "spider91" \? "抓取" : "扫盘"\}/);
+  assert.match(driveComponentsSource, /label=\{d\.kind === "spider91" \? "已废弃" : "扫盘"\}/);
  assert.match(driveComponentsSource, /status=\{d\.scanGenerationStatus\}/);
  assert.match(driveComponentsSource, /showCounts=\{false\}/);
-  assert.match(driveComponentsSource, /label === "抓取" && state === "scanning" \? "抓取中"/);
  assert.match(driveComponentsSource, /status\?\.scannedCount/);
  assert.match(driveComponentsSource, /预计新增/);
  assert.match(apiSource, /scannedCount:\s*number/);
@@ -241,6 +286,13 @@ test("drive generation panel shows scan or crawler status first", () => {
  assert.match(constantsSource, /if \(state === "scanning"\) return "扫盘中"/);
 });

+test("legacy spider91 storage is disabled in drive management", () => {
+  assert.match(drivesPageSource, /91Spider 不再支持通过网盘运行，请到爬虫管理添加爬虫脚本/);
+  assert.match(drivesPageSource, /disabled=\{d\.kind === "spider91"\}/);
+  assert.match(drivesPageSource, /已废弃，请到爬虫管理添加/);
+  assert.match(constantsSource, /91Spider 不再支持通过网盘添加或编辑/);
+});
+
 test("drive detail selection is stored in the URL history", () => {
  assert.match(drivesPageSource, /useSearchParams/);
  assert.match(drivesPageSource, /searchParams\.get\("drive"\)/);