diff --git a/app/api/endpoints/douban.py b/app/api/endpoints/douban.py
index fa8e9cd0f..0bfc2cdb8 100644
--- a/app/api/endpoints/douban.py
+++ b/app/api/endpoints/douban.py
@@ -68,9 +68,9 @@ async def douban_credits(
     count: CompatibleCountParam = None,
 ) -> Any:
     """
-    根据豆瓣ID查询演员阵容，type_name: 电影/电视剧
+    根据豆瓣ID查询演员阵容，type_name: 电影/movie、电视剧/tv
     """
-    mediatype = MediaType(type_name)
+    mediatype = MediaType.from_agent(type_name) or MediaType(type_name)
     if mediatype == MediaType.MOVIE:
         return await DoubanChain().async_movie_credits(doubanid=doubanid)
     elif mediatype == MediaType.TV:
@@ -91,9 +91,9 @@ async def douban_recommend(
     _: _SchemaTokenPayload = Depends(verify_token),
 ) -> Any:
     """
-    根据豆瓣ID查询推荐电影/电视剧，type_name: 电影/电视剧
+    根据豆瓣ID查询推荐电影/电视剧，type_name: 电影/movie、电视剧/tv
     """
-    mediatype = MediaType(type_name)
+    mediatype = MediaType.from_agent(type_name) or MediaType(type_name)
     if mediatype == MediaType.MOVIE:
         medias = await DoubanChain().async_movie_recommend(doubanid=doubanid)
     elif mediatype == MediaType.TV:
diff --git a/app/api/endpoints/media.py b/app/api/endpoints/media.py
index fc256483c..ef1e67c4f 100644
--- a/app/api/endpoints/media.py
+++ b/app/api/endpoints/media.py
@@ -697,7 +697,7 @@ async def detail(
     """
     根据媒体来源和原生 ID 查询影视或音乐信息
     """
-    mtype = MediaType(type_name)
+    mtype = MediaType.from_agent(type_name) or MediaType(type_name)
     normalized_source, normalized_media_id = resolve_media_identity(
         media_source=media_source,
         media_id=media_id,
diff --git a/app/api/endpoints/mediaserver.py b/app/api/endpoints/mediaserver.py
index dc20fce75..231106dc8 100644
--- a/app/api/endpoints/mediaserver.py
+++ b/app/api/endpoints/mediaserver.py
@@ -162,11 +162,11 @@ def not_exists(
     count: CompatibleCountParam = None,
 ) -> Any:
     """
-    根据媒体信息查询缺失电影/剧集
+    根据媒体信息查询缺失电影/剧集，媒体类型兼容英文与中文值
     """
     # 媒体信息
     meta = MetaInfo(title=media_in.title)
-    mtype = MediaType(media_in.type) if media_in.type else None
+    mtype = MediaType.from_agent(media_in.type) or MediaType(media_in.type) if media_in.type else None
     if mtype:
         meta.type = mtype
     if media_in.season is not None:
@@ -176,7 +176,8 @@ def not_exists(
         meta.year = media_in.year
     # 转化为媒体信息对象
     mediainfo = MediaInfo()
-    mediainfo.from_dict(media_in.model_dump())
+    # 复用已解析的类型，避免领域对象再次将 Agent 英文值按中文枚举解析。
+    mediainfo.from_dict({**media_in.model_dump(), "type": mtype})
     exist_flag, no_exists = DownloadChain().get_no_exists_info(
         meta=meta, mediainfo=mediainfo
     )
diff --git a/app/api/endpoints/subscribe.py b/app/api/endpoints/subscribe.py
index 9e4148698..41d3bc42d 100644
--- a/app/api/endpoints/subscribe.py
+++ b/app/api/endpoints/subscribe.py
@@ -251,7 +251,7 @@ async def create_subscribe(
     """
     # 类型转换
     if subscribe_in.type:
-        mtype = MediaType(subscribe_in.type)
+        mtype = MediaType.from_agent(subscribe_in.type) or MediaType(subscribe_in.type)
     else:
         mtype = None
     # 非 TMDB 来源的标题可能自带季标记，入库前统一拆分。
@@ -625,7 +625,7 @@ async def popular_subscribes(
             if min_sub and count < min_sub:
                 continue
             media = MediaInfo()
-            media.type = MediaType(sub.get("type"))
+            media.type = MediaType.from_agent(sub.get("type")) or MediaType(sub.get("type"))
             media.media_source = normalize_media_source(sub.get("media_source"))
             media.media_id = str(sub.get("media_id")) if sub.get("media_id") is not None else None
             # 处理标题
diff --git a/app/api/endpoints/tmdb.py b/app/api/endpoints/tmdb.py
index b51a87069..88ff98dea 100644
--- a/app/api/endpoints/tmdb.py
+++ b/app/api/endpoints/tmdb.py
@@ -114,9 +114,9 @@ async def tmdb_similar(
     _: _SchemaTokenPayload = Depends(verify_token),
 ) -> Any:
     """
-    根据TMDBID查询类似电影/电视剧，type_name: 电影/电视剧
+    根据TMDBID查询类似电影/电视剧，type_name: 电影/movie、电视剧/tv
     """
-    mediatype = MediaType(type_name)
+    mediatype = MediaType.from_agent(type_name) or MediaType(type_name)
     if mediatype == MediaType.MOVIE:
         medias = await TmdbChain().async_movie_similar(tmdbid=tmdbid)
     elif mediatype == MediaType.TV:
@@ -141,9 +141,9 @@ async def tmdb_recommend(
     _: _SchemaTokenPayload = Depends(verify_token),
 ) -> Any:
     """
-    根据TMDBID查询推荐电影/电视剧，type_name: 电影/电视剧
+    根据TMDBID查询推荐电影/电视剧，type_name: 电影/movie、电视剧/tv
     """
-    mediatype = MediaType(type_name)
+    mediatype = MediaType.from_agent(type_name) or MediaType(type_name)
     if mediatype == MediaType.MOVIE:
         medias = await TmdbChain().async_movie_recommend(tmdbid=tmdbid)
     elif mediatype == MediaType.TV:
@@ -187,9 +187,9 @@ async def tmdb_credits(
     _: _SchemaTokenPayload = Depends(verify_token),
 ) -> Any:
     """
-    根据TMDBID查询演员阵容，type_name: 电影/电视剧
+    根据TMDBID查询演员阵容，type_name: 电影/movie、电视剧/tv
     """
-    mediatype = MediaType(type_name)
+    mediatype = MediaType.from_agent(type_name) or MediaType(type_name)
     if mediatype == MediaType.MOVIE:
         persons = await TmdbChain().async_movie_credits(tmdbid=tmdbid, page=page)
     elif mediatype == MediaType.TV:
diff --git a/app/api/endpoints/transfer.py b/app/api/endpoints/transfer.py
index ebe6a9182..b423d6b1a 100644
--- a/app/api/endpoints/transfer.py
+++ b/app/api/endpoints/transfer.py
@@ -790,8 +790,9 @@ def _execute_manual_transfer(
     mtype = None
     type_name = str(transer_item.type_name).strip() if transer_item.type_name else ""
     if type_name and type_name.lower() not in {"自动", "auto", "none"}:
+        # 先按 Agent 使用的 movie/tv/music 取值解析，再回退到原生中文枚举值。
         try:
-            mtype = MediaType(type_name)
+            mtype = MediaType.from_agent(type_name) or MediaType(type_name)
         except ValueError:
             return _SchemaResponse(
                 success=False, message=f"不支持的媒体类型：{type_name}"
diff --git a/docs/mcp-api.md b/docs/mcp-api.md
index fef758174..aa2296707 100644
--- a/docs/mcp-api.md
+++ b/docs/mcp-api.md
@@ -419,6 +419,8 @@ FastAPI 的 HTTP 异常和参数校验异常统一使用 `message`，不再返
 
 媒体来源列表 `/api/v1/media/source` 仅预置上述九个来源，其余来源由启用插件注册后提供。哔哩哔哩、芒果 TV、咪咕视频、腾讯视频、爱奇艺不再占用内置来源标识，宿主也不再转换这些插件来源的旧别名；调用方应使用插件声明的准确来源 ID。
 
+媒体详情的 `type_name`、新增订阅和媒体库缺失查询的 `type`、手动整理的 `type_name` 均兼容 `movie`、`tv`、`music` 及对应中文值 `电影`、`电视剧`、`音乐`。TMDB 的相似、推荐和演员接口，以及豆瓣的推荐和演员接口同样兼容英文类型，但仍只提供电影和电视剧结果；传入音乐类型返回空列表。
+
 影视自动识别在未指定来源时只使用 TMDB，未命中时不会继续查询其它影视源。音乐路径识别严格按 AcoustID 音频指纹、文件标签、文件名三级依次执行；指纹或标签直接提供 MusicBrainz Recording ID 时，会直接查询 MusicBrainz 详情，未绑定身份的音乐标题及目录识别按 `SEARCH_SOURCE` 中内置音乐来源的配置顺序回退（MusicBrainz、TheAudioDB、豆瓣音乐），未选择音乐来源时兼容默认 MusicBrainz。显式 `media_source` / 主身份或 MusicBrainz 发行标签固定所属来源；来源之间不转换或拼接 ID。`MediaInfo` 响应仍可能包含 `tmdb_id`、`douban_id`、`bangumi_id`、`anilist_id` 等跨源映射辅助字段，但这些字段不是通用请求入口。明确归属 `/tmdb`、`/douban`、`/bangumi`、`/anilist` 的接口，以及固定使用 TMDB 的剧集组和排期接口，仍可按其单数据源契约接收原生 ID。
 
 | 方法 | 路径 | 说明 |
diff --git a/skills/moviepilot-api/SKILL.md b/skills/moviepilot-api/SKILL.md
index 24624e656..ae91fb3df 100644
--- a/skills/moviepilot-api/SKILL.md
+++ b/skills/moviepilot-api/SKILL.md
@@ -1,6 +1,6 @@
 ---
 name: moviepilot-api
-version: 48
+version: 49
 description: >-
   Use this skill for MoviePilot product operations such as media search, torrent
   search, downloads, subscriptions, library checks, sites, storage, workflows,
@@ -169,6 +169,10 @@ are separate operation contracts, not interchangeable parameter templates.
 subscription executions use `limit`, not `page` or `count`. `site.rss` lists
 RSS-enabled sites; it does not accept a `site_id` filter.
 
+`media.detail` and `transfer.file` accept `movie`, `tv`, or `music` in
+`type_name`; `subscription.add` accepts the same values in `type`. The
+corresponding Chinese labels remain supported.
+
 ## API Category Index
 
 Each category file contains the complete operation contracts for its namespace.
diff --git a/tests/test_api_media_types.py b/tests/test_api_media_types.py
new file mode 100644
index 000000000..0b635710e
--- /dev/null
+++ b/tests/test_api_media_types.py
@@ -0,0 +1,160 @@
+"""验证 API 将 Agent 英文类型与原生中文类型传入相同业务分支。"""
+
+import asyncio
+from types import SimpleNamespace
+from unittest.mock import AsyncMock, Mock
+
+import pytest
+
+from app.api.endpoints import douban, media, mediaserver, subscribe, tmdb, transfer
+from app.schemas.file import FileItem
+from app.schemas.subscribe import Subscribe
+from app.schemas.transfer import ManualTransferItem
+from app.schemas.types import MediaSource, MediaType
+from app.schemas.workflow import MediaInfo
+
+MEDIA_TYPES = [
+    ("movie", MediaType.MOVIE),
+    ("tv", MediaType.TV),
+    ("music", MediaType.MUSIC),
+    ("电影", MediaType.MOVIE),
+    ("电视剧", MediaType.TV),
+    ("音乐", MediaType.MUSIC),
+    (" MOVIE ", MediaType.MOVIE),
+    (" TV ", MediaType.TV),
+    (" MUSIC ", MediaType.MUSIC),
+]
+METADATA_ENDPOINTS = [
+    (douban, "DoubanChain", "douban_credits", "credits", "doubanid"),
+    (douban, "DoubanChain", "douban_recommend", "recommend", "doubanid"),
+    (tmdb, "TmdbChain", "tmdb_similar", "similar", "tmdbid"),
+    (tmdb, "TmdbChain", "tmdb_recommend", "recommend", "tmdbid"),
+    (tmdb, "TmdbChain", "tmdb_credits", "credits", "tmdbid"),
+]
+
+
+@pytest.mark.parametrize("type_name,expected", MEDIA_TYPES)
+@pytest.mark.parametrize("module,chain_name,endpoint,operation,id_key", METADATA_ENDPOINTS)
+def test_metadata_endpoints_route_media_types(
+    monkeypatch, type_name, expected, module, chain_name, endpoint, operation, id_key,
+) -> None:
+    """影视英文和中文类型选择相同方法，音乐类型保持空列表。"""
+    movie_call = AsyncMock(return_value=[])
+    tv_call = AsyncMock(return_value=[])
+    chain = SimpleNamespace(**{
+        f"async_movie_{operation}": movie_call,
+        f"async_tv_{operation}": tv_call,
+    })
+    monkeypatch.setattr(module, chain_name, lambda: chain)
+
+    result = asyncio.run(getattr(module, endpoint)(
+        **{id_key: "123" if id_key == "doubanid" else 123},
+        type_name=type_name,
+    ))
+
+    assert result == []
+    assert movie_call.await_count == int(expected == MediaType.MOVIE)
+    assert tv_call.await_count == int(expected == MediaType.TV)
+
+
+@pytest.mark.parametrize("type_name,expected", MEDIA_TYPES)
+def test_media_detail_forwards_normalized_type(monkeypatch, type_name, expected) -> None:
+    """详情入口应将三类中英文类型都转换成枚举再交给识别链。"""
+    recognize = AsyncMock(return_value=None)
+    monkeypatch.setattr(media, "MediaChain", lambda: SimpleNamespace(
+        async_recognize_media=recognize,
+    ))
+
+    asyncio.run(media.detail(
+        media_id="123", media_source=MediaSource.TMDB, type_name=type_name,
+    ))
+
+    recognize.assert_awaited_once_with(
+        media_source=MediaSource.TMDB, media_id="123", mtype=expected,
+    )
+
+
+@pytest.mark.parametrize("type_name,expected", MEDIA_TYPES)
+def test_create_subscription_forwards_normalized_type(monkeypatch, type_name, expected) -> None:
+    """新增订阅应兼容 Agent 类型，并向订阅链传入准确枚举。"""
+    add = AsyncMock(return_value=(1, ""))
+    monkeypatch.setattr(subscribe, "SubscribeChain", lambda: SimpleNamespace(async_add=add))
+
+    result = asyncio.run(subscribe.create_subscribe(
+        subscribe_in=Subscribe(name="测试媒体", type=type_name),
+        current_user=SimpleNamespace(name="admin", is_superuser=True),
+    ))
+
+    assert result.success is True
+    assert add.await_args.kwargs["mtype"] is expected
+
+
+@pytest.mark.parametrize("type_name,expected", MEDIA_TYPES)
+def test_popular_subscriptions_normalize_record_type(monkeypatch, type_name, expected) -> None:
+    """共享订阅记录使用英文类型时，公开响应仍输出原生中文类型。"""
+    statistic = AsyncMock(return_value=[{
+        "name": "测试媒体", "type": type_name, "count": 1,
+    }])
+    monkeypatch.setattr(
+        subscribe.MoviePilotServerHelper, "async_get_subscribe_statistic", statistic,
+    )
+
+    result = asyncio.run(subscribe.popular_subscribes(stype="all"))
+
+    assert len(result) == 1
+    assert result[0]["type"] == expected.value
+
+
+@pytest.mark.parametrize("type_name,expected", MEDIA_TYPES + [(None, MediaType.UNKNOWN)])
+def test_library_missing_forwards_meta_type(monkeypatch, type_name, expected) -> None:
+    """缺失查询应向下载链传入解析后的类型，空值沿用标题推断。"""
+    missing = Mock(return_value=(True, {}))
+    monkeypatch.setattr(mediaserver, "DownloadChain", lambda: SimpleNamespace(
+        get_no_exists_info=missing,
+    ))
+
+    assert mediaserver.not_exists(media_in=MediaInfo(title="测试电影", type=type_name)) == []
+    assert missing.call_args.kwargs["meta"].type is expected
+    assert missing.call_args.kwargs["mediainfo"].type is (
+        expected if type_name is not None else None
+    )
+
+
+@pytest.mark.parametrize("type_name,expected", MEDIA_TYPES + [
+    (None, None), ("", None), ("自动", None), ("auto", None), ("none", None),
+])
+def test_manual_transfer_forwards_normalized_type(monkeypatch, type_name, expected) -> None:
+    """手动整理兼容中英文类型，并保留未指定和自动识别语义。"""
+    execute = Mock(return_value=(True, ""))
+    monkeypatch.setattr(transfer, "TransferChain", lambda: SimpleNamespace(manual_transfer=execute))
+
+    result = transfer.manual_transfer(
+        transer_item=ManualTransferItem(
+            fileitem=FileItem(storage="local", path="/downloads/test.mkv", type="file"),
+            type_name=type_name,
+        ),
+        background=True,
+        history_query=Mock(),
+    )
+
+    assert result.success is True
+    assert execute.call_args.kwargs["mtype"] is expected
+
+
+def test_manual_transfer_rejects_invalid_type_before_execution(monkeypatch) -> None:
+    """非法媒体类型应保持失败响应，不能启动实际整理。"""
+    chain_factory = Mock()
+    monkeypatch.setattr(transfer, "TransferChain", chain_factory)
+
+    result = transfer.manual_transfer(
+        transer_item=ManualTransferItem(
+            fileitem=FileItem(storage="local", path="/downloads/test.mkv", type="file"),
+            type_name="unsupported",
+        ),
+        background=True,
+        history_query=Mock(),
+    )
+
+    assert result.success is False
+    assert result.message == "不支持的媒体类型：unsupported"
+    chain_factory.assert_not_called()
diff --git a/tests/test_builtin_skill_boundaries.py b/tests/test_builtin_skill_boundaries.py
index 3449e0569..1c4201d24 100644
--- a/tests/test_builtin_skill_boundaries.py
+++ b/tests/test_builtin_skill_boundaries.py
@@ -117,7 +117,7 @@ def test_modified_builtin_skills_have_incremented_versions() -> None:
         "command-dispatch": "2",
         "database-operation": "9",
         "feedback-issue": "11",
-        "moviepilot-api": "48",
+        "moviepilot-api": "49",
         "moviepilot-update": "5",
         "organize-files": "5",
         "transfer-failed-retry": "5",