diff --git a/app/api/endpoints/download.py b/app/api/endpoints/download.py
index cdc2df0ed..d2f191c2b 100644
--- a/app/api/endpoints/download.py
+++ b/app/api/endpoints/download.py
@@ -18,6 +18,7 @@ from app.api.response import ResponseAPIRouter
 from app.chain.download import DownloadChain
 from app.chain.media import MediaChain
 from app.domain.context import Context, MediaInfo, MusicInfo, SubtitleInfo, TorrentInfo
+from app.domain.meta.metabase import MetaBase
 from app.domain.meta.metamusic import MetaMusic
 from app.domain.metainfo import MetaInfo
 from app.adapters.web.security.access import verify_token
@@ -71,6 +72,44 @@ def _prepare_subtitle_download(
     return True, ""
 
 
+def _build_unrecognized_media_info(
+    torrent: _SchemaTorrentInfo,
+    metainfo: MetaBase,
+    is_music: bool = False,
+    music_type: Optional[str] = None,
+) -> MediaInfo | MusicInfo:
+    """
+    为用户确认的未识别资源构造最小下载上下文，影视与音乐统一处理。
+
+    影视以种子分类兜底媒体类型并保留标题年份，音乐按解析标题构造音乐信息，
+    两者都不再要求识别出统一媒体信息即可继续下载。
+    """
+    if is_music:
+        return MusicInfo(
+            title=metainfo.title or torrent.title,
+            year=metainfo.year,
+            music_type=music_type or MUSIC_ENTITY_RECORDING,
+        )
+    try:
+        media_type = MediaType(torrent.category)
+    except (TypeError, ValueError):
+        media_type = MediaType.from_agent(torrent.category)
+    if media_type == MediaType.COLLECTION:
+        media_type = MediaType.MOVIE
+    if media_type not in (MediaType.MOVIE, MediaType.TV):
+        media_type = metainfo.type
+        # 合集类型在回退到元数据后同样归一为电影，避免落到 UNKNOWN
+        if media_type == MediaType.COLLECTION:
+            media_type = MediaType.MOVIE
+    if media_type not in (MediaType.MOVIE, MediaType.TV):
+        media_type = MediaType.UNKNOWN
+    return MediaInfo(
+        type=media_type,
+        title=metainfo.name or torrent.title,
+        year=metainfo.year,
+    )
+
+
 @router.get("/", summary="正在下载", response_model=List[_SchemaDownloaderTorrent])
 def current(
     name: Optional[str] = None, _: _SchemaTokenPayload = Depends(verify_token)
@@ -134,6 +173,7 @@ def add(
     media_source: Annotated[MediaSource | None, Body()] = None,
     media_id: Annotated[str | None, Body()] = None,
     music_type: Annotated[MusicTargetEntityType | None, Body()] = None,
+    allow_unrecognized: Annotated[bool, Body()] = False,
     downloader: Annotated[str | None, Body()] = None,
     # 保存路径, 支持<storage>:<path>, 如rclone:/MP, smb:/server/share/Movies等
     save_path: Annotated[str | None, Body()] = None,
@@ -189,7 +229,19 @@ def add(
             music_type=normalized_music_type,
         )
     if not mediainfo:
-        return _SchemaResponse(success=False, message="无法识别媒体信息")
+        if not allow_unrecognized:
+            return _SchemaResponse(
+                success=False,
+                message="无法识别媒体信息",
+                data=_SchemaDownloadAddedData(requires_confirmation=True),
+            )
+        # 用户已确认：影视与音乐统一按种子元信息构造最小上下文继续下载
+        mediainfo = _build_unrecognized_media_info(
+            torrent_in,
+            metainfo,
+            is_music=is_music,
+            music_type=normalized_music_type,
+        )
     # 种子信息
     torrentinfo = TorrentInfo()
     torrentinfo.from_dict(torrent_in.model_dump())
diff --git a/app/schemas/download.py b/app/schemas/download.py
index 59f21a605..ec09d0b12 100644
--- a/app/schemas/download.py
+++ b/app/schemas/download.py
@@ -31,6 +31,10 @@ class DownloadAddedData(BaseModel):
     """下载任务添加结果。"""
 
     download_id: Optional[str] = Field(default=None, description="下载任务 ID")
+    requires_confirmation: bool = Field(
+        default=False,
+        description="是否需要用户确认后下载未识别资源",
+    )
 
 
 class SubtitleDownloadData(BaseModel):
diff --git a/docs/mcp-api.md b/docs/mcp-api.md
index 337eea787..df5a7a3de 100644
--- a/docs/mcp-api.md
+++ b/docs/mcp-api.md
@@ -215,7 +215,7 @@ AniList 榜单、探索、详情、人物和推荐接口优先通过 `anilist-ch
 | :--- | :--- | :--- |
 | GET | `/api/v1/download/` | 查询正在下载的任务，参数：`name`；关联下载历史时返回媒体类型、来源站点 `site_name`，以及 `media.poster` 海报和 `media.backdrop` 背景图；兼容字段 `media.image` 与 `media.poster` 相同 |
 | POST | `/api/v1/download/` | 添加含媒体信息的下载任务，请求体包含媒体信息和种子信息 |
-| POST | `/api/v1/download/add` | 添加不含媒体信息的下载任务，请求体包含 `torrent_in`，可选且必须成对提供 `media_source` + `media_id`，并支持 `music_type`、`downloader`、`save_path` |
+| POST | `/api/v1/download/add` | 添加不含媒体信息的下载任务，请求体包含 `torrent_in`，可选且必须成对提供 `media_source` + `media_id`，并支持 `music_type`、`downloader`、`save_path`；影视或音乐识别失败时统一响应 `data.requires_confirmation=true`，用户确认后可用 `allow_unrecognized=true` 重试本次下载 |
 | POST | `/api/v1/download/subtitle` | 下载字幕到识别出的媒体下载目录，请求体包含 `subtitle_in`，并必须提供 `media_source` + `media_id`；可选 `save_path` |
 | GET | `/api/v1/download/start/{hashString}` | 恢复下载任务，参数：`name` |
 | GET | `/api/v1/download/stop/{hashString}` | 暂停下载任务，参数：`name` |
diff --git a/skills/moviepilot-api/SKILL.md b/skills/moviepilot-api/SKILL.md
index 94a353ff9..dbdee70d7 100644
--- a/skills/moviepilot-api/SKILL.md
+++ b/skills/moviepilot-api/SKILL.md
@@ -235,7 +235,7 @@ Streaming search sends `{"type":"heartbeat"}` every 15 seconds without business
 |--------|------|-------------|
 | GET | `/api/v1/download/` | List active downloads. Params: `name` (downloader name); linked history adds media type and source `site_name` |
 | POST | `/api/v1/download/` | Add download (with media info). Body: JSON |
-| POST | `/api/v1/download/add` | Add download without media info. Body: `torrent_in`, optional paired `media_source` + `media_id`, `music_type`, `downloader`, `save_path` |
+| POST | `/api/v1/download/add` | Add download without media info. Body: `torrent_in`, optional paired `media_source` + `media_id`, `music_type`, `downloader`, `save_path`; an unrecognized video or music resource returns `data.requires_confirmation=true`, and the same request may be retried with `allow_unrecognized=true` after explicit user confirmation |
 | POST | `/api/v1/download/subtitle` | Download subtitle file to the recognized media download directory. Body: `subtitle_in`, required `media_source` + `media_id`, optional `save_path` |
 | GET | `/api/v1/download/start/{hashString}` | Resume download task |
 | GET | `/api/v1/download/stop/{hashString}` | Pause download task |
diff --git a/tests/test_download_media_source.py b/tests/test_download_media_source.py
index 4ac6e9b3a..9a3394fd9 100644
--- a/tests/test_download_media_source.py
+++ b/tests/test_download_media_source.py
@@ -1,4 +1,5 @@
 from types import SimpleNamespace
+from unittest.mock import Mock
 
 from app import schemas
 from app.api.endpoints import download as download_endpoint
@@ -61,6 +62,102 @@ def test_download_add_rejects_source_without_media_id() -> None:
     assert response.message == "媒体来源和媒体 ID 必须同时提供"
 
 
+def test_download_add_requires_confirmation_when_recognition_fails(monkeypatch) -> None:
+    """未识别的影视资源必须先由用户确认，不能直接提交下载。"""
+    media_chain = Mock()
+    media_chain.recognize_by_meta.return_value = None
+    download_chain = Mock()
+    monkeypatch.setattr(download_endpoint, "MediaChain", lambda: media_chain)
+    monkeypatch.setattr(download_endpoint, "DownloadChain", lambda: download_chain)
+
+    response = download_endpoint.add(
+        torrent_in=schemas.TorrentInfo(
+            title="Harry Potter Complete Collection",
+            category=MediaType.MOVIE.value,
+        ),
+        current_user=SimpleNamespace(name="tester"),
+    )
+
+    assert response.success is False
+    assert response.message == "无法识别媒体信息"
+    assert response.data.requires_confirmation is True
+    download_chain.download_single.assert_not_called()
+
+
+def test_download_add_allows_confirmed_unrecognized_video(monkeypatch) -> None:
+    """用户确认后应使用种子元数据提交未识别的影视合集。"""
+    media_chain = Mock()
+    media_chain.recognize_by_meta.return_value = None
+    download_chain = Mock()
+    download_chain.download_single.return_value = "download-collection"
+    monkeypatch.setattr(download_endpoint, "MediaChain", lambda: media_chain)
+    monkeypatch.setattr(download_endpoint, "DownloadChain", lambda: download_chain)
+
+    response = download_endpoint.add(
+        torrent_in=schemas.TorrentInfo(
+            title="Harry Potter Complete Collection",
+            category="movie",
+        ),
+        allow_unrecognized=True,
+        current_user=SimpleNamespace(name="tester"),
+    )
+
+    assert response.success is True
+    context = download_chain.download_single.call_args.kwargs["context"]
+    assert context.media_info.type == MediaType.MOVIE
+    assert context.media_info.title == "Harry Potter Complete Collection"
+    assert context.media_info.media_id is None
+
+
+def test_download_add_requires_confirmation_for_unrecognized_music(monkeypatch) -> None:
+    """未识别的音乐资源同样需要先由用户确认，不能直接提交下载。"""
+    media_chain = Mock()
+    media_chain.recognize_by_meta.return_value = None
+    download_chain = Mock()
+    monkeypatch.setattr(download_endpoint, "MediaChain", lambda: media_chain)
+    monkeypatch.setattr(download_endpoint, "DownloadChain", lambda: download_chain)
+
+    response = download_endpoint.add(
+        torrent_in=schemas.TorrentInfo(
+            title="Various Artists - 90s Collection",
+            category=MediaType.MUSIC.value,
+        ),
+        music_type="album",
+        current_user=SimpleNamespace(name="tester"),
+    )
+
+    assert response.success is False
+    assert response.message == "无法识别媒体信息"
+    assert response.data.requires_confirmation is True
+    download_chain.download_single.assert_not_called()
+
+
+def test_download_add_allows_confirmed_unrecognized_music(monkeypatch) -> None:
+    """用户确认后音乐资源也应使用种子元数据继续下载。"""
+    media_chain = Mock()
+    media_chain.recognize_by_meta.return_value = None
+    download_chain = Mock()
+    download_chain.download_single.return_value = "download-music"
+    monkeypatch.setattr(download_endpoint, "MediaChain", lambda: media_chain)
+    monkeypatch.setattr(download_endpoint, "DownloadChain", lambda: download_chain)
+
+    response = download_endpoint.add(
+        torrent_in=schemas.TorrentInfo(
+            title="Various Artists - 90s Collection",
+            category="music",
+        ),
+        music_type="album",
+        allow_unrecognized=True,
+        current_user=SimpleNamespace(name="tester"),
+    )
+
+    assert response.success is True
+    context = download_chain.download_single.call_args.kwargs["context"]
+    assert context.media_info.type == MediaType.MUSIC
+    assert context.media_info.music_type == "album"
+    assert context.media_info.title == "90s Collection"
+
+
 def test_subtitle_download_passes_generic_media_source(monkeypatch) -> None:
     """字幕下载接口应把统一来源ID传递到下载链。"""
     captured = {}