diff --git a/parser/src/main/resources/py/feishu-dl.py b/parser/src/main/resources/py/feishu-dl.py
new file mode 100644
index 0000000..0127449
--- /dev/null
+++ b/parser/src/main/resources/py/feishu-dl.py
@@ -0,0 +1,355 @@
+#!/usr/bin/env python3
+"""
+飞书公开分享 直链解析 + 批量下载 (aria2/Motrix)
+支持: 单文件链接 / 文件夹链接(递归子目录)
+
+用法:
+  python feishu_dl.py <链接>                     # 推送到 Motrix
+  python feishu_dl.py <链接> -d D:/Downloads      # 指定下载目录
+  python feishu_dl.py <链接> --list               # 仅列出文件,不下载
+  python feishu_dl.py <链接> --aria2c              # 输出 aria2c 命令行
+"""
+
+import sys, os, re, json, uuid, ssl, gzip, argparse
+import http.cookiejar
+import urllib.request, urllib.error
+from urllib.parse import unquote, quote
+
+# ─── Motrix aria2 RPC 默认配置 ──────────────────────────
+ARIA2_RPC_URL = "http://localhost:16800/jsonrpc"
+ARIA2_SECRET  = "motrix"
+# ────────────────────────────────────────────────────────
+
+UA = ("Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
+      "(KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36")
+
+# 飞书 obj_type 映射 (type=12 上传文件可下载, type=0 文件夹可递归)
+OBJ_TYPES = {
+    0: "📁 文件夹", 2: "📝 旧版文档", 3: "📊 表格", 8: "🧠 思维导图",
+    11: "📽 幻灯片", 12: "📄 文件", 22: "📝 新版文档", 30: "📋 画板",
+    44: "📊 多维表格", 84: "📑 知识库", 123: "❓ 未知", 124: "❓ 未知",
+}
+
+# v3 列表 API 支持的 obj_type
+LIST_OBJ_TYPES = [0, 2, 22, 44, 3, 30, 8, 11, 12, 84, 123, 124]
+
+
+# ─── 网络工具 ────────────────────────────────────────────
+
+def _ctx():
+    ctx = ssl.create_default_context()
+    ctx.check_hostname = False
+    ctx.verify_mode = ssl.CERT_NONE
+    return ctx
+
+def make_opener(jar):
+    return urllib.request.build_opener(
+        urllib.request.HTTPSHandler(context=_ctx()),
+        urllib.request.HTTPCookieProcessor(jar),
+    )
+
+def decode_body(resp):
+    data = resp.read()
+    if resp.headers.get("Content-Encoding") == "gzip":
+        data = gzip.decompress(data)
+    return data.decode("utf-8", errors="replace")
+
+def cookie_string(jar):
+    return "; ".join(f"{c.name}={c.value}" for c in jar)
+
+def human_size(n):
+    for u in ("B", "KB", "MB", "GB"):
+        if n < 1024: return f"{n:.1f} {u}"
+        n /= 1024
+    return f"{n:.1f} TB"
+
+
+# ─── 飞书核心 API ────────────────────────────────────────
+
+def parse_share_url(url):
+    """返回 (tenant, token, link_type:'file'|'folder')"""
+    m = re.match(r'https://([^.]+)\.feishu\.cn/file/([A-Za-z0-9_-]+)', url)
+    if m: return m.group(1), m.group(2), "file"
+    m = re.match(r'https://([^.]+)\.feishu\.cn/drive/folder/([A-Za-z0-9_-]+)', url)
+    if m: return m.group(1), m.group(2), "folder"
+    return None, None, None
+
+
+def fetch_session(share_url):
+    """访问分享页拿匿名 session cookie"""
+    jar = http.cookiejar.CookieJar()
+    opener = make_opener(jar)
+    req = urllib.request.Request(share_url)
+    req.add_header("User-Agent", UA)
+    req.add_header("Accept", "text/html,*/*")
+    opener.open(req, timeout=15).read()
+    return jar
+
+
+def list_folder(tenant, folder_token, jar, page_label=""):
+    """
+    v3 API 列出文件夹内容 (单页)
+    GET /space/api/explorer/v3/children/list/?token=xxx&length=50&...
+    """
+    base = f"https://{tenant}.feishu.cn"
+    params = ["length=50", "asc=1", "rank=5", f"token={folder_token}"]
+    for t in LIST_OBJ_TYPES:
+        params.append(f"obj_type={t}")
+    if page_label:
+        params.append(f"last_label={quote(page_label, safe='')}")
+
+    url = f"{base}/space/api/explorer/v3/children/list/?{'&'.join(params)}"
+    opener = make_opener(jar)
+    req = urllib.request.Request(url)
+    req.add_header("User-Agent", UA)
+    req.add_header("Accept", "application/json, text/plain, */*")
+    req.add_header("Referer", f"{base}/drive/folder/{folder_token}")
+
+    resp = opener.open(req, timeout=15)
+    data = json.loads(decode_body(resp))
+    if data.get("code") != 0:
+        raise RuntimeError(f"API error: {data.get('msg')}")
+
+    d = data["data"]
+    nodes = d.get("entities", {}).get("nodes", {})
+    node_list = d.get("node_list", [])
+
+    items = []
+    for nid in node_list:
+        node = nodes.get(nid, {})
+        obj_type = node.get("type", -1)
+        obj_token = node.get("obj_token", "")
+        name = node.get("name", "unknown")
+        extra = node.get("extra", {})
+        try:   size = int(extra.get("size", "0"))
+        except: size = 0
+        items.append({
+            "name": name, "obj_token": obj_token, "type": obj_type,
+            "size": size, "url": node.get("url", ""),
+            "is_folder": obj_type == 0,
+            "type_name": OBJ_TYPES.get(obj_type, f"❓ type={obj_type}"),
+        })
+
+    # 排除文件夹自身节点
+    items = [it for it in items if it["obj_token"] != folder_token]
+    return items, d.get("has_more", False), d.get("last_label", "")
+
+
+def list_folder_all(tenant, folder_token, jar):
+    """分页获取文件夹全部内容"""
+    all_items, label = [], ""
+    while True:
+        items, has_more, label = list_folder(tenant, folder_token, jar, label)
+        all_items.extend(items)
+        if not has_more: break
+    return all_items
+
+
+def walk_folder(tenant, folder_token, jar, prefix="", depth=0):
+    """递归遍历, 返回扁平列表 [{..., path:"a/b/file.txt"}]"""
+    if depth > 10:  # 防止无限递归
+        return []
+    items = list_folder_all(tenant, folder_token, jar)
+    result = []
+    for it in items:
+        if it["is_folder"]:
+            sub = walk_folder(tenant, it["obj_token"], jar,
+                              prefix=f"{prefix}{it['name']}/", depth=depth+1)
+            result.extend(sub)
+        else:
+            it["path"] = f"{prefix}{it['name']}"
+            result.append(it)
+    return result
+
+
+def probe_file(tenant, obj_token, jar, referer):
+    """Range 探测文件名 + 大小 (只取1字节)"""
+    dl_url = f"https://{tenant}.feishu.cn/space/api/box/stream/download/all/{obj_token}"
+    opener = make_opener(jar)
+    req = urllib.request.Request(dl_url)
+    req.add_header("User-Agent", UA)
+    req.add_header("Referer", referer)
+    req.add_header("Range", "bytes=0-0")
+
+    resp = opener.open(req, timeout=15)
+    cd = resp.headers.get("Content-Disposition", "")
+    cr = resp.headers.get("Content-Range", "")
+    resp.read()
+
+    filename = ""
+    m = re.search(r"filename\*=UTF-8''(.+?)(?:;|$)", cd)
+    if m: filename = unquote(m.group(1).strip())
+    if not filename:
+        m = re.search(r'filename="?([^";]+)"?', cd)
+        if m: filename = unquote(m.group(1).strip())
+
+    total = 0
+    m = re.search(r'/(\d+)', cr)
+    if m: total = int(m.group(1))
+    return filename, total
+
+
+# ─── aria2 RPC ───────────────────────────────────────────
+
+def aria2_add(dl_url, cs, referer, filename, out_dir=None):
+    opts = {"header": [f"Cookie: {cs}", f"Referer: {referer}", f"User-Agent: {UA}"]}
+    if filename: opts["out"] = filename
+    if out_dir:  opts["dir"] = out_dir
+    payload = json.dumps({
+        "jsonrpc": "2.0", "id": str(uuid.uuid4()),
+        "method": "aria2.addUri",
+        "params": [f"token:{ARIA2_SECRET}", [dl_url], opts],
+    }).encode()
+    req = urllib.request.Request(ARIA2_RPC_URL, data=payload,
+                                 headers={"Content-Type": "application/json"})
+    resp = urllib.request.urlopen(req, timeout=10)
+    return json.loads(resp.read().decode()).get("result", "")
+
+
+def push_one(dl_url, cs, referer, filename, out_dir, quiet=False):
+    try:
+        gid = aria2_add(dl_url, cs, referer, filename, out_dir)
+        if gid:
+            if not quiet: print(f"      [✓] GID={gid}  {filename}")
+            return True
+    except urllib.error.URLError:
+        if not quiet:
+            print(f"      [✗] Motrix 未启动, RPC: {ARIA2_RPC_URL}")
+    except Exception as e:
+        if not quiet: print(f"      [✗] {e}")
+    return False
+
+
+def print_aria2c(dl_url, cs, referer, filename, out_dir):
+    print(f'aria2c --header="Cookie: {cs}" \\')
+    print(f'  --header="Referer: {referer}" \\')
+    print(f'  --header="User-Agent: {UA}" \\')
+    if filename: print(f'  -o "{filename}" \\')
+    if out_dir:  print(f'  -d "{out_dir}" \\')
+    print(f'  "{dl_url}"')
+
+
+# ─── 主流程 ──────────────────────────────────────────────
+
+def handle_file(tenant, token, jar, args):
+    share_url = f"https://{tenant}.feishu.cn/file/{token}"
+    dl_url = f"https://{tenant}.feishu.cn/space/api/box/stream/download/all/{token}"
+    cs = cookie_string(jar)
+
+    print(f"[2/3] 探测文件 ...")
+    filename, size = probe_file(tenant, token, jar, share_url)
+    print(f"      {filename}  ({human_size(size)})")
+
+    if args.list: return
+    if args.aria2c:
+        print_aria2c(dl_url, cs, share_url, filename, args.dir); return
+
+    print(f"[3/3] 推送到 Motrix ...")
+    if not push_one(dl_url, cs, share_url, filename, args.dir):
+        print(f"\n  降级输出 aria2c 命令:\n")
+        print_aria2c(dl_url, cs, share_url, filename, args.dir)
+
+
+def handle_folder(tenant, token, jar, args):
+    base = f"https://{tenant}.feishu.cn"
+    cs = cookie_string(jar)
+
+    print(f"[2/4] 递归扫描文件夹 ...")
+    all_files = walk_folder(tenant, token, jar)
+
+    downloadable = [f for f in all_files if f["type"] == 12]
+    skipped      = [f for f in all_files if f["type"] != 12]
+
+    print(f"      共 {len(all_files)} 项: "
+          f"{len(downloadable)} 可下载, {len(skipped)} 在线文档(跳过)")
+
+    if skipped:
+        print(f"\n  ⏭ 跳过的在线文档:")
+        for f in skipped:
+            print(f"    {f['type_name']}  {f['path']}")
+
+    if not downloadable:
+        print("\n  没有可下载的文件"); return
+
+    # 探测真实文件名和大小
+    print(f"\n[3/4] 探测文件信息 ...")
+    total_size = 0
+    for f in downloadable:
+        referer = f.get("url", f"{base}/drive/folder/{token}")
+        try:
+            real_name, size = probe_file(tenant, f["obj_token"], jar, referer)
+            f["real_name"] = real_name or f["name"]
+            f["size"] = size or f["size"]
+        except:
+            f["real_name"] = f["name"]
+        total_size += f["size"]
+
+    # 打印文件列表
+    print(f"\n  {'─'*60}")
+    print(f"  {'#':>3}  {'文件名':<35} {'大小':>10}  路径")
+    print(f"  {'─'*60}")
+    for i, f in enumerate(downloadable):
+        sz = human_size(f["size"]) if f["size"] else "  ?"
+        print(f"  {i+1:>3}  {f['real_name']:<35} {sz:>10}  {f['path']}")
+    print(f"  {'─'*60}")
+    print(f"  合计: {len(downloadable)} 文件, {human_size(total_size)}")
+
+    if args.list: return
+
+    # 下载
+    print(f"\n[4/4] {'aria2c 命令' if args.aria2c else '推送 Motrix'} ...")
+    ok = 0
+    for f in downloadable:
+        dl_url = f"{base}/space/api/box/stream/download/all/{f['obj_token']}"
+        sub = os.path.dirname(f["path"])
+        out_dir = os.path.join(args.dir, sub) if args.dir and sub else (args.dir or None)
+
+        if args.aria2c:
+            print_aria2c(dl_url, cs, base, f["real_name"], out_dir)
+            print()
+            ok += 1
+        else:
+            if push_one(dl_url, cs, base, f["real_name"], out_dir, quiet=True):
+                ok += 1
+                print(f"      [✓] {f['real_name']}")
+            else:
+                print(f"      [✗] {f['real_name']} — Motrix 未响应")
+                print(f"          降级 aria2c:\n")
+                print_aria2c(dl_url, cs, base, f["real_name"], out_dir)
+                return  # Motrix 挂了就不继续了
+
+    print(f"\n      完成! {ok}/{len(downloadable)}")
+
+
+# ─── 入口 ────────────────────────────────────────────────
+
+def main():
+    ap = argparse.ArgumentParser(description="飞书分享解析下载器 v2")
+    ap.add_argument("url", help="飞书分享链接 (文件/文件夹)")
+    ap.add_argument("-d", "--dir", default=None, help="下载目录")
+    ap.add_argument("--list", action="store_true", help="仅列出,不下载")
+    ap.add_argument("--aria2c", action="store_true", help="输出 aria2c 命令")
+    args = ap.parse_args()
+
+    tenant, token, lt = parse_share_url(args.url)
+    if not token:
+        print(f"[✗] 不支持的链接: {args.url}")
+        print(f"    格式: https://xxx.feishu.cn/file/xxxToken")
+        print(f"          https://xxx.feishu.cn/drive/folder/xxxToken")
+        sys.exit(1)
+
+    print(f"\n{'📄' if lt=='file' else '📁'} 飞书分享解析  [{lt}]  {tenant}/{token}")
+
+    print(f"[1/{3 if lt=='file' else 4}] 获取匿名会话 ...")
+    jar = fetch_session(args.url)
+    print(f"      {sum(1 for _ in jar)} cookies")
+
+    if lt == "file":
+        handle_file(tenant, token, jar, args)
+    else:
+        handle_folder(tenant, token, jar, args)
+    print()
+
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file