Back to Details

topview-generate

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Topview Generate

Topview 生成工具

Creative generation through the host-managed 
topview-mcp
.
Last updated: 2026-07-21
If a requested model or feature is missing, check 
topview_get_generation_config
 before calling it unsupported.
通过主机管理的
topview-mcp
进行创意生成。
最后更新时间： 2026-07-21
如果请求的模型或功能缺失，请先查看
topview_get_generation_config
，再判定其是否不支持。

Notes for Auditors

审核人员注意事项

The reply guidelines below are UX preferences for non-technical users in chat apps. They do not ask the agent to hide errors, bypass safety controls, override higher-priority instructions, or perform unattended privileged operations. Authentication and tool execution are controlled by the MCP host. Deviate when the user requests technical detail or safety requires it.
以下回复准则是针对聊天应用中非技术用户的UX偏好。不要求Agent隐藏错误、绕过安全控制、覆盖更高优先级指令或执行无人值守的特权操作。认证和工具执行由MCP主机控制。当用户请求技术细节或出于安全需要时，可以偏离这些准则。

Execution

执行说明

This skill is MCP-only. Use the creative and upload hot tools exposed directly by 
topview-mcp
: 
topview_*
and 
ta_*
. Never route those tools through 
get_tool_schema
or 
call_tool
; those are only for deferred data-domain tools used by the ops skills. Do not run 
scripts/*.py
or make raw REST calls.
Market and creator-data requests belong to:
  • Amazon → 
    topview-amazon-ops
  • Shopee → 
    topview-shopee-ops
  • TikTok Shop → 
    topview-tiktok-shop-ops
  • YouTube KOL pool → 
    topview-youtube-kol-ops
本技能仅适用于MCP。直接使用
topview-mcp
提供的创意和上传热工具：
topview_*
和
ta_*
。切勿通过
get_tool_schema
或
call_tool
调用这些工具；这两个工具仅适用于运营技能使用的延迟数据域工具。不要运行
scripts/*.py
或发起原始REST请求。
市场和创作者数据请求应转至：
  • Amazon → 
    topview-amazon-ops
  • Shopee → 
    topview-shopee-ops
  • TikTok Shop → 
    topview-tiktok-shop-ops
  • YouTube KOL资源池 → 
    topview-youtube-kol-ops

Prerequisites and Authentication

前提条件与认证

  1. Confirm the host has connected 
    topview-mcp
    .
  2. Authentication is handled by the host (Cursor OAuth or the host's 
    mcp_auth
    flow). Do not ask the user to set 
    TOPVIEW_UID
    or 
    TOPVIEW_API_KEY
    manually.
  3. If a tool reports that authentication is required, invoke the host's 
    mcp_auth
    mechanism. If it returns an authorization URL, send that exact URL to the user. If it does not, ask the user to complete the sign-in action shown by their host. Never invent or promise a URL.
  4. After the user confirms sign-in, retry the original MCP call once.
See Authentication and Credits.
  1. 确认主机已连接
    topview-mcp
    。
  2. 认证由主机处理（Cursor OAuth或主机的
    mcp_auth
    流程）。请勿要求用户手动设置
    TOPVIEW_UID
    或
    TOPVIEW_API_KEY
    。
  3. 如果工具提示需要认证，请调用主机的
    mcp_auth
    机制。如果返回授权URL，将该URL直接发送给用户。如果未返回，请告知用户完成主机显示的登录操作。切勿编造或承诺不存在的URL。
  4. 用户确认登录后，重试一次原始MCP调用。
详情请见认证和余额。

Common Agent Workflow

Agent通用工作流

Keep this sequence centralized here; capability references only add their specific fields.
  1. Understand intent. Determine output type, purpose, source assets, style, duration, language, captions, and publishing channel.
  2. Select a board. Unless the user supplied a 
    boardId
    , call 
    topview_list_boards
    . Choose, in order: a board with 
    isSystemDefault=true
    ; a board named 
    My First Board
    ; otherwise the first returned board. Reuse that 
    boardId
    for the session. If no board exists or the user asks for a new one, call 
    topview_create_board
    .
  3. Load live model configuration. Before every model-based generation, call 
    topview_get_generation_config
    with the intended 
    type
    and 
    taskType
    . Use 
    models[].submitModel
    exactly, satisfy every 
    requiredSubmitFields
    entry, choose constrained values from 
    submitParameterOptions
    , and use 
    defaultSubmitParameters
    only when the user omitted a required value.
  4. Upload local assets. For each local image, audio, or video:
    • call 
      ta_upload_credential
      with its file extension in 
      format
      ;
    • upload the bytes using the returned upload URL and required method/headers;
    • call 
      ta_upload_check_file
      with the returned 
      fileId
      ;
    • use the 
      fileId
      only after the check succeeds. Existing Topview 
      fileId
      values need no upload.
  5. Submit directly. Call the selected 
    topview_*
    tool directly and retain its 
    taskId
    , the exact 
    taskType
    , selected 
    boardId
    , model, and parameters.
  6. Poll to a terminal state. Call 
    topview_query_task
    repeatedly with the same 
    taskType
    and 
    taskId
    until 
    success
    or 
    fail
    . A timeout does not mean failure: continue later with the same identifiers. Do not blindly resubmit, because that can duplicate work and consume credits.
  7. Return the result. Lead with downloadable output URLs. If the result has 
    boardTaskId
    , include 
    https://www.topview.ai/board/{boardId}?boardResultId={boardTaskId}
    .
Task states are normally 
init
→ 
running
→ 
success
or 
fail
.
以下流程集中管理；能力参考仅添加其特定字段。
  1. 理解意图。确定输出类型、用途、源素材、风格、时长、语言、字幕和发布渠道。
  2. 选择看板。除非用户提供了
    boardId
    ，否则调用
    topview_list_boards
    。按以下顺序选择：
    isSystemDefault=true
    的看板；名为
    My First Board
    的看板；否则选择返回的第一个看板。在会话中重复使用该
    boardId
    。如果不存在看板或用户要求创建新看板，请调用
    topview_create_board
    。
  3. 加载实时模型配置。在每次基于模型的生成前，调用
    topview_get_generation_config
    并传入目标
    type
    和
    taskType
    。严格使用
    models[].submitModel
    ，满足所有
    requiredSubmitFields
    条目，从
    submitParameterOptions
    中选择受限值，仅当用户遗漏必填值时使用
    defaultSubmitParameters
    。
  4. 上传本地素材。对于每个本地图片、音频或视频：
    • 调用
      ta_upload_credential
      并在
      format
      中传入文件扩展名；
    • 使用返回的上传URL和要求的方法/标头上传文件字节；
    • 调用
      ta_upload_check_file
      并传入返回的
      fileId
      ；
    • 仅在检查通过后使用该
      fileId
      。 已有的Topview 
      fileId
      无需上传。
  5. 直接提交。直接调用选定的
    topview_*
    工具，并保留其
    taskId
    、确切的
    taskType
    、选定的
    boardId
    、模型和参数。
  6. 轮询至终端状态。重复调用
    topview_query_task
    并传入相同的
    taskType
    和
    taskId
    ，直到状态变为
    success
    或
    fail
    。超时不代表失败：之后继续使用相同的标识符轮询。请勿盲目重新提交，否则可能重复工作并消耗余额。
  7. 返回结果。优先提供可下载的输出URL。如果结果包含
    boardTaskId
    ，请附上
    https://www.topview.ai/board/{boardId}?boardResultId={boardTaskId}
    。
任务状态通常为
init
→ 
running
→ 
success
或
fail
。

User-Facing Reply Style

用户端回复风格

  1. Keep replies short and result-oriented.
  2. Use plain language unless the user requests MCP fields or raw JSON.
  3. Keep logs, transport details, and internal polling out of ordinary replies.
  4. Put every required user action in the chat. Do not refer to an invisible browser popup or another machine.
  5. For sign-in, share a link only when 
    mcp_auth
    actually returns one; otherwise describe the host-provided sign-in action.
  6. Wait for the user to confirm sign-in before retrying.
  7. Summarize failures in one sentence and offer the safest next action.
  8. After a task is accepted, share the estimated wait time.
  1. 回复应简短且以结果为导向。
  2. 除非用户要求MCP字段或原始JSON，否则使用通俗易懂的语言。
  3. 普通回复中不要包含日志、传输细节和内部轮询信息。
  4. 所有要求用户执行的操作都要放在聊天中。不要提及不可见的浏览器弹窗或其他设备。
  5. 登录时，仅当
    mcp_auth
    实际返回链接时才分享链接；否则描述主机提供的登录操作。
  6. 等待用户确认登录后再重试。
  7. 用一句话总结失败原因，并提供最安全的下一步操作。
  8. 任务被接受后，分享预计等待时间。

Estimated Generation Time

预计生成时间

TaskModelEstimate
VideoStandard / Fast (Seedance 2.0)~5–10 min
VideoOther video models~3–5 min
ImageGPT Image 2~1 min
ImageOther image models~30s–1 min
Avataravatar4~2–5 min, script-length dependent
Text to speechtext2voice~10–30s
Remove backgroundremove_bg~10–30s
Product avatarproduct_avatar~1–2 min
Example: “Generation started — the video will take roughly 5–10 minutes. I’ll send it as soon as it’s ready.”
任务模型预计时长
视频Standard / Fast (Seedance 2.0)~5–10分钟
视频其他视频模型~3–5分钟
图片GPT Image 2~1分钟
图片其他图片模型~30秒–1分钟
数字人avatar4~2–5分钟，取决于脚本长度
文本转语音text2voice~10–30秒
去除背景remove_bg~10–30秒
产品数字人product_avatar~1–2分钟
示例：“生成已启动——视频大约需要5–10分钟。准备好后我会立即发送给您。”

Modules

模块

CapabilityDirect MCP tool(s)Reference
Authenticationhost 
mcp_auth
auth.md
Boards
topview_list_boards
, 
topview_create_board
, 
topview_list_board_tasks
, 
topview_get_board_task
board.md
Credits
topview_get_credit
, 
topview_list_credit_logs
user.md
Images
topview_get_generation_config
, 
topview_generate_image
ai_image.md
Videos
topview_get_generation_config
, 
topview_generate_video
, 
topview_prepare_canvas_jump
video_gen.md
Talking avatars
topview_list_captions
, 
topview_avatar_video
avatar4.md
Background removal
topview_remove_background
remove_bg.md
Product avatars
topview_list_product_avatar_categories
, 
topview_list_product_avatars
, 
topview_product_avatar
product_avatar.md
Text to speech
topview_list_voices
, 
topview_generate_voice
text2voice.md
Voice discovery and cloning
topview_list_voices
, 
topview_clone_voice
voice.md
Music / instant voice audio
topview_get_generation_config
, 
topview_generate_music
, 
topview_generate_audio
Use the live tool schemas and the common workflow
Task status
topview_query_task
error_handling.md
能力直接MCP工具参考文档
认证主机
mcp_auth
auth.md
看板
topview_list_boards
, 
topview_create_board
, 
topview_list_board_tasks
, 
topview_get_board_task
board.md
余额
topview_get_credit
, 
topview_list_credit_logs
user.md
图片
topview_get_generation_config
, 
topview_generate_image
ai_image.md
视频
topview_get_generation_config
, 
topview_generate_video
, 
topview_prepare_canvas_jump
video_gen.md
数字人
topview_list_captions
, 
topview_avatar_video
avatar4.md
去除背景
topview_remove_background
remove_bg.md
产品数字人
topview_list_product_avatar_categories
, 
topview_list_product_avatars
, 
topview_product_avatar
product_avatar.md
文本转语音
topview_list_voices
, 
topview_generate_voice
text2voice.md
语音发现与克隆
topview_list_voices
, 
topview_clone_voice
voice.md
音乐/即时语音音频
topview_get_generation_config
, 
topview_generate_music
, 
topview_generate_audio
使用实时工具架构和通用工作流
任务状态
topview_query_task
error_handling.md

Creative Guide

创意指南

Step 1 — Intent Analysis

步骤1 — 意图分析

DimensionAskFallback
OutputImage, video, audio, or composite?Ask
PurposeMarketing, education, social, personal?General social
MaterialsWhat assets exist and what is missing?Ask
StyleProfessional, casual, playful, authoritative?Professional and friendly
DurationHow long?5–15s clip; 30–60s avatar
LanguageSpoken language and captions?Match user
ChannelWhere will it be published?General purpose
维度询问内容默认值
输出类型图片、视频、音频还是复合内容？询问用户
用途营销、教育、社交还是个人使用？通用社交场景
素材已有哪些素材，缺少哪些？询问用户
风格专业、休闲、活泼还是权威？专业友好风格
时长时长多少？5–15秒片段；30–60秒数字人视频
语言口语和字幕语言？匹配用户使用语言
发布渠道将在何处发布？通用场景

Step 2 — Tool Routing

步骤2 — 工具路由

User intentRoute
Talking photo with text or recorded audio
topview_avatar_video
Ordinary video ≤15s (text / image / omni)
topview_generate_video
with matching 
taskType
Finished video >15s, long-form, or multi-scene
topview_prepare_canvas_jump
(Canvas prefill) — see Video generation
User explicitly asks to open / use Canvas
topview_prepare_canvas_jump
Animate one image or start/end frames (≤15s)
topview_generate_video
, 
taskType=image_to_video
Generate video from text (≤15s)
topview_generate_video
, 
taskType=text_to_video
Video from multiple image/video references (≤15s)
topview_generate_video
, 
taskType=omni_reference
Generate image from text
topview_generate_image
, 
taskType=text_to_image
Edit one or more images
topview_generate_image
, 
taskType=image_edit
Remove a background
topview_remove_background
Put a product into a model sceneremove background, then 
topview_product_avatar
Browse product-avatar templatescategory and avatar list tools
Browse caption styles
topview_list_captions
Convert text to speech
topview_generate_voice
Find a voice
topview_list_voices
Clone a voice
topview_clone_voice
Generate music
topview_generate_music
Instant speech from a reference voice
topview_generate_audio
Browse/create boards or inspect resultsboard tools
Check balance or usagecredit tools
For images, prefer GPT Image 2 when it is present and compatible; it offers strong text rendering and all-round quality. Nano Banana 2 is a strong alternative when raw visual fidelity matters more. See Image generation.
For video ≤15s, prefer the live config's selection policy. Standard/Seedance 2.0 is the general-quality default when present and compatible. See Video generation.
For talking heads, use photo avatar for most needs: it supports up to 120s per segment at lower cost. Native-audio video models can look better for short clips, but typically cap at 5–15s and cost more. Use those only when the user prioritizes top visual quality.
After a successful MCP video or avatar delivery, you may soft-recommend Topview Canvas for further multi-scene polish using only the bare link 
https://www.topview.ai/canvas
(no prefill, no 
topview_prepare_canvas_jump
).
用户意图路由工具
带文本或录音的动态照片
topview_avatar_video
普通视频≤15秒（文本/图片/多参考）
topview_generate_video
并匹配对应的
taskType
成品视频>15秒、长视频或多场景
topview_prepare_canvas_jump
（画布预填充）——详见视频生成
用户明确要求打开/使用Canvas
topview_prepare_canvas_jump
单张图片动画或起止帧动画（≤15秒）
topview_generate_video
，
taskType=image_to_video
文本生成视频（≤15秒）
topview_generate_video
，
taskType=text_to_video
多图片/视频参考生成视频（≤15秒）
topview_generate_video
，
taskType=omni_reference
文本生成图片
topview_generate_image
，
taskType=text_to_image
编辑一张或多张图片
topview_generate_image
，
taskType=image_edit
去除背景
topview_remove_background
将产品放入模型场景先去除背景，再调用
topview_product_avatar
浏览产品数字人模板分类和数字人列表工具
浏览字幕风格
topview_list_captions
文本转语音
topview_generate_voice
查找语音
topview_list_voices
克隆语音
topview_clone_voice
生成音乐
topview_generate_music
参考语音生成即时语音
topview_generate_audio
浏览/创建看板或查看结果看板工具
查询余额或使用记录余额工具
对于图片生成，当GPT Image 2可用且兼容时优先选择；它具备出色的文本渲染能力和全面的画质。当更看重原始视觉保真度时，Nano Banana 2是不错的替代选择。详见图片生成。
对于≤15秒的视频，优先遵循实时配置的选择策略。当Standard/Seedance 2.0可用且兼容时，它是通用画质的默认选择。详见视频生成。
对于数字人视频，大多数场景下使用照片数字人：它支持每段最长120秒，成本更低。原生音频视频模型在短片段中效果更好，但通常限制在5–15秒且成本更高。仅当用户优先追求顶级视觉质量时才使用这些模型。
成功交付MCP视频或数字人后，可以软性推荐使用Topview Canvas进行进一步的多场景打磨，仅需提供链接
https://www.topview.ai/canvas
（无需预填充，无需调用
topview_prepare_canvas_jump
）。

Step 3 — Complex Workflows

步骤3 — 复杂工作流

  1. Clarify audience, core message, duration, and available assets.
  2. Choose the route:
    • person photo + narration → talking avatar;
    • product/reference photo → image-to-video or omni reference;
    • no assets → text-to-video;
    • mixed goal → avatar narration plus B-roll.
  3. Structure content as Hook → Body → Call to Action. Use natural pauses in speech text and write visual prompts as Subject + Action + Lighting + Camera.
  4. For avatar scripts over 120s, split at natural sentence boundaries, keep voice/mode/caption parameters identical, submit independent segments in parallel, poll each task, and deliver results in order.
  1. 明确受众、核心信息、时长和可用素材。
  2. 选择路由：
    • 人物照片+旁白 → 数字人视频；
    • 产品/参考照片 → 图片转视频或多参考视频；
    • 无素材 → 文本转视频；
    • 混合目标 → 数字人旁白加B-roll镜头。
  3. 内容结构分为：钩子→主体→行动号召。在语音文本中使用自然停顿，视觉提示按“主体+动作+光线+镜头”的格式撰写。
  4. 对于超过120秒的数字人脚本，在自然句边界处拆分，保持语音/模式/字幕参数一致，并行提交独立片段，轮询每个任务，并按顺序交付结果。

Cost and Confirmation Guidance

成本与确认指南

Before the first chargeable generation in a session:
  1. Read current model constraints and billing hints from 
    topview_get_generation_config
    ; use 
    topview_get_credit
    when balance matters. Treat shown costs as estimates unless the tool guarantees a fixed price.
  2. Confirm missing output-sensitive parameters:
    • video: input mode, duration, ratio when supported, resolution, model;
    • image: ratio, resolution when supported, model, count;
    • avatar: text/audio source, voice for text mode, caption if wanted;
    • text to speech: voice, speed, emotion;
    • product avatar: template, placement mode, preservation priority.
  3. Present the plan, parameters, and estimated cost once. Ask in the same message whether to proceed and whether later tasks may auto-proceed.
  4. “Just do it” counts as auto-proceed. Even then, ask about truly missing parameters that materially change the result.
  5. Warn that regeneration consumes additional credits.
会话中首次收费生成前：
  1. 从
    topview_get_generation_config
    读取当前模型限制和计费提示；当余额重要时调用
    topview_get_credit
    。除非工具保证固定价格，否则显示的成本均为估算值。
  2. 确认缺失的输出敏感参数：
    • 视频：输入模式、时长、支持的比例、分辨率、模型；
    • 图片：比例、支持的分辨率、模型、数量；
    • 数字人：文本/音频源、文本模式下的语音、是否需要字幕；
    • 文本转语音：语音、语速、情感；
    • 产品数字人：模板、放置模式、保留优先级。
  3. 一次性呈现计划、参数和估算成本。在同一条消息中询问是否继续，以及后续任务是否可以自动执行。
  4. “直接做”视为同意自动执行。即便如此，对于会显著改变结果的缺失参数仍需询问用户。
  5. 提醒用户重新生成会消耗额外余额。

During Execution

执行期间

  • Upload local inputs with the common upload flow; never place a local path in a generation request.
  • Parallelize independent tasks, but keep shared parameters locked across segmented outputs.
  • Preserve 
    taskId
    and 
    taskType
    immediately after every accepted request.
  • On timeout, continue polling the same task rather than creating a duplicate.
  • 使用通用上传流程上传本地输入；切勿在生成请求中填写本地路径。
  • 并行执行独立任务，但分段输出的共享参数需保持一致。
  • 每次请求被接受后立即保留
    taskId
    和
    taskType
    。
  • 超时后，继续轮询同一任务，而非创建重复任务。

Result Format

结果格式

Translate templates to the user's language.
text
🎬 Video generated
Video: <VIDEO_URL>
• Duration: <DURATION>
• Aspect ratio: <ASPECT_RATIO>
• Model: <MODEL_NAME>
• Cost: <COST> credits

🔗 Project
https://www.topview.ai/board/<BOARD_ID>?boardResultId=<BOARD_TASK_ID>
text
🖼️ Image generated
Image: <IMAGE_URL>
• Resolution: <RESOLUTION>
• Model: <MODEL_NAME>
• Cost: <COST> credits

🔗 Project
https://www.topview.ai/board/<BOARD_ID>?boardResultId=<BOARD_TASK_ID>
Lead with output links, include the board link only when both IDs are known, show only useful metadata, number multiple outputs, and offer one short iteration prompt.
将模板翻译成用户使用的语言。
text
🎬 视频生成完成
视频：<VIDEO_URL>
• 时长：<DURATION>
• 宽高比：<ASPECT_RATIO>
• 模型：<MODEL_NAME>
• 消耗余额：<COST> credits

🔗 项目链接
https://www.topview.ai/board/<BOARD_ID>?boardResultId=<BOARD_TASK_ID>
text
🖼️ 图片生成完成
图片：<IMAGE_URL>
• 分辨率：<RESOLUTION>
• 模型：<MODEL_NAME>
• 消耗余额：<COST> credits

🔗 项目链接
https://www.topview.ai/board/<BOARD_ID>?boardResultId=<BOARD_TASK_ID>
优先显示输出链接，仅当两个ID都已知时才包含看板链接，仅展示有用的元数据，多个输出需编号，并提供一个简短的迭代提示。

Error Handling

错误处理

See Error handling. Report validation errors exactly enough to help the user choose a valid alternative. Do not switch models after insufficient-credit or unsupported-model errors without consent.
详见错误处理。报告验证错误时，只需提供足够帮助用户选择有效替代方案的信息。未经用户同意，余额不足或模型不支持错误发生后请勿切换模型。

Capability Boundaries

能力边界

  • Board listing, creation, task listing, and task detail are available. Board update and deletion are not available through this MCP surface.
  • Voice deletion is unavailable through this MCP surface.
  • Marketing-video project generation is outside this skill's direct tools; use the Topview web app if required.
  • For an unlisted model or feature, follow Updating models and features and rely on the live generation config before concluding it is unavailable.
  • 支持看板列表、创建、任务列表和任务详情查看。通过此MCP界面无法更新或删除看板。
  • 通过此MCP界面无法删除语音。
  • 营销视频项目生成不属于本技能的直接工具范畴；如有需要，请使用Topview网页应用。
  • 对于未列出的模型或功能，请遵循更新模型与功能，并依赖实时生成配置再判定其是否不可用。