run-featured-workflow

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Featured Workflows (run-workflow MCP)

精选工作流（run-workflow MCP）

Run pre-built marketing and production workflows bundled with the MCP server. Never compose from scratch when

get_featured_workflow

finds a match.

运行MCP服务器自带的预构建营销与生产工作流。当

get_featured_workflow

找到匹配项时，无需从头构建工作流。

When to use

使用场景

TRIGGER when the user:

Names a featured workflow: retargeting, localization, banner advertising, packaging, product banners at scale, catalogs, packshots, spin video, etc.
Asks to run a "known" or "predefined" workflow by name or keyword
Provides product images + fonts for a banner/merge-data workflow with defaults

DO NOT TRIGGER when:

User wants a custom workflow with no named match — use run-workflow and
```
compose_workflow
```
User pastes only a batch/execution/workflow ID — use
```
inspect_run
```
via run-workflow skill

满足以下条件时触发：

用户指定某个精选工作流：retargeting、localization、横幅广告、包装设计、批量产品横幅、商品目录、产品实拍图、旋转视频等
用户要求按名称或关键词运行“已知”或“预定义”工作流
用户提供产品图片+字体，用于使用默认配置的横幅/数据合并工作流

以下情况请勿触发：

用户需要无匹配命名的自定义工作流——请使用run-workflow和
```
compose_workflow
```
用户仅粘贴批量/执行/工作流ID——请通过run-workflow skill调用
```
inspect_run
```

Tool routing

工具路由

Step	Tool
Find predefined workflow	`get_featured_workflow`
Upload images, fonts, custom templates, CSV data	`upload_asset`
Extract merge tags from custom `.indd`	`get_indesign_tags`
Submit (large / multi-banner)	`run_workflow_submit`
Poll progress and outputs	`run_workflow_get_status` (always `includeOutputs: true` )
Re-wire graph after tag diff	`compose_workflow` (custom templates only)
Diagnose failure	`inspect_run`

步骤	工具
查找预定义工作流	`get_featured_workflow`
上传图片、字体、自定义模板、CSV数据	`upload_asset`
从自定义 `.indd` 文件提取合并标签	`get_indesign_tags`
提交（大型/多横幅任务）	`run_workflow_submit`
查询进度与输出结果	`run_workflow_get_status` （需始终设置 `includeOutputs: true` ）
标签差异后重新关联工作流图	`compose_workflow` （仅适用于自定义模板）
诊断失败原因	`inspect_run`

Discovery flow

发现流程

Call

get_featured_workflow({ query: "<keyword>" })

Multiple matches —
```
prepared
```
is absent. List matched names, ask user to pick, retry with exact name.
Single match — present
```
mergeNodes
```
, required inputs, sizes. Confirm with user.
Ask about templates if not specified (defaults vs custom
```
.indd
```
).
Ask about fonts if user has
```
.otf
```
/
```
.ttf
```
files — upload via
```
upload_asset
```
.
If the user supplies a
```
.csv
```
(data-merge workflows like
```
product-banners-at-scale
```
), upload it and wire it to the input-files
node (
```
type: "file"
```
in
```
prepared.inputNodes
```
) — see Data/CSV inputs below.

Distinct asset types route to distinct input nodes: images →

input-images

, fonts →

defaults.fonts

, CSV/data →

input-files

(which feeds

parse-data

). Never collapse them onto one node.

调用

get_featured_workflow({ query: "<关键词>" })

多个匹配项——返回结果中无
```
prepared
```
字段。列出所有匹配的工作流名称，询问用户选择后，使用精确名称重试调用
单个匹配项——展示
```
mergeNodes
```
、必填输入项和尺寸，与用户确认
若未指定模板，询问用户使用默认模板还是自定义
```
.indd
```
模板
若用户拥有
```
.otf
```
/
```
.ttf
```
字体文件，询问是否需要通过
```
upload_asset
```
上传
若用户提供
```
.csv
```
文件（适用于
```
product-banners-at-scale
```
这类数据合并工作流），上传文件并将其关联到**
```
input-files
```
节点**（对应
```
prepared.inputNodes
```
中的
```
type: "file"
```
）——详见下方「数据/CSV输入」部分

**不同类型的资产需关联到不同输入节点：**图片→

input-images

，字体→

defaults.fonts

，CSV/数据→

input-files

（关联至

parse-data

节点）。切勿将所有资产合并到单个节点。

Fast path (defaults)

快速路径（默认配置）

When user wants default templates and default text:

Upload product images and fonts via
```
upload_asset
```
.
Call
```
run_workflow_submit
```
:

json

{
  "session_id": "<prepared.session_id>",
  "useSessionDefaults": true,
  "defaults": {
    "fonts": [{ "presignedUrl": "...", "name": "Font.otf", "storageType": "azure" }]
  },
  "inputs": [
    {
      "node_id": "<input-images id from prepared.inputNodes>",
      "content": [{ "type": "image", "url": "..." }]
    }
  ]
}

Do not call
```
compose_workflow
```
.
Do not send
```
actions
```
/
```
connections
```
when
```
session_id
```
is present.
Pass all images in one
```
content
```
array — one submit call.

当用户希望使用默认模板且默认文本时：

通过
```
upload_asset
```
上传产品图片和字体
调用
```
run_workflow_submit
```
：

json

{
  "session_id": "<prepared.session_id>",
  "useSessionDefaults": true,
  "defaults": {
    "fonts": [{ "presignedUrl": "...", "name": "Font.otf", "storageType": "azure" }]
  },
  "inputs": [
    {
      "node_id": "<prepared.inputNodes中的input-images节点ID>",
      "content": [{ "type": "image", "url": "..." }]
    }
  ]
}

请勿调用
```
compose_workflow
```
当
```
session_id
```
存在时，请勿发送
```
actions
```
/
```
connections
```
参数
将所有图片放入同一个
```
content
```
数组——只需一次提交调用

Data / CSV inputs

数据/CSV输入

Some featured workflows (e.g.

product-banners-at-scale

) have a CSV branch:

input-files

→

parse-data

→

merge-data

prepared.inputNodes

surfaces the CSV node with

type: "file"

, and

prepared.inputs

carries the workflow's baked sample CSV as inline content.

Defaults only (no user CSV): submit as-is — the baked sample CSV flows through, no
```
input-files
```
entry needed.
User-supplied CSV: upload it via
```
upload_asset
```
, then add an
```
inputs[]
```
entry for the
```
input-files
```
node id so it overrides the baked default:

json

{
  "node_id": "<input-files id from prepared.inputNodes>",
  "content": [{ "url": "<uploaded csv url>", "mimeType": "text/csv" }]
}

The CSV columns drive the

parse-data

fan-out (e.g.

headline

subcopy

CTA

) — do not also wire those as separate text inputs. A product image that's constant across every row (not a per-row CSV column) wires

input-images

directly into

merge-data

, bypassing

parse-data

entirely — this is the privileged default; only route an image through

parse-data

's media-override port when it genuinely varies per row.

部分精选工作流（如

product-banners-at-scale

）包含CSV分支：

input-files

→

parse-data

→

merge-data

。

prepared.inputNodes

会展示类型为

file

的CSV节点，

prepared.inputs

则包含工作流内置的示例CSV作为内嵌内容。

**仅使用默认配置（无用户CSV）：**直接提交即可——内置示例CSV会自动流转，无需添加
```
input-files
```
条目
**用户提供CSV：**通过
```
upload_asset
```
上传文件，然后为
```
input-files
```
节点ID添加
```
inputs[]
```
条目，以覆盖内置默认值：

json

{
  "node_id": "<prepared.inputNodes中的input-files节点ID>",
  "content": [{ "url": "<已上传的CSV文件链接>", "mimeType": "text/csv" }]
}

CSV列会驱动

parse-data

节点的输出（如

headline

、

subcopy

、

CTA

）——请勿将这些列作为单独的文本输入关联。若某张产品图片在所有行中保持不变（非每行CSV列的值），则直接将

input-images

关联至

merge-data

节点，完全绕过

parse-data

节点——这是默认的最优配置；只有当图片确实需要逐行变化时，才将其通过

parse-data

节点的媒体覆盖端口进行关联。

Custom template path

自定义模板路径

When user provides custom

.indd

templates:

```
get_indesign_tags
```
on each template.
Compare to
```
prepared.mergeNodes[].tags
```
:
- Identical → swap template URLs only; skip compose.
- Different → feasibility check, then optional
```
compose_workflow
```
  re-wire.
If a critical image tag was removed (e.g.
```
background
```
in retargeting), stop — offer defaults, different template, or build from scratch.
Show updated graph visually. Prompt for missing merge-data inputs only.
Submit after user confirms all values.

Merge-data vs scene prompts: Only InDesign merge-port tags need user-facing text/image inputs. Keep

gen-object-composite

scene prompts at their defaults unless user asks to change them.

当用户提供自定义

.indd

模板时：

对每个模板调用
```
get_indesign_tags
```
与
```
prepared.mergeNodes[].tags
```
进行对比：
- 标签完全一致——仅替换模板链接；跳过
```
compose_workflow
```
  调用
- 标签不一致——先进行可行性检查，再选择性调用
```
compose_workflow
```
  重新关联工作流
若关键图片标签被移除（如retargeting工作流中的
```
background
```
标签），则终止流程——为用户提供使用默认模板、更换其他模板或从头构建的选项
可视化展示更新后的工作流图。仅当缺少数据合并输入项时，提示用户补充
用户确认所有值后提交任务

**数据合并与场景提示：**仅InDesign合并端口标签需要用户提供文本/图片输入。除非用户要求修改，否则保持

gen-object-composite

的场景提示为默认值。

Polling and outputs

轮询与输出结果

After

run_workflow_submit

, poll

run_workflow_get_status

every ~5 seconds with

includeOutputs: true

Report:

Running — N/M complete (X%) · Ns elapsed

using

elapsedSeconds

from response.

On completion: list every output's full presigned URL verbatim. Never truncate.
On failure:
```
inspect_run
```
, summarize, stop — do not auto-retry.

调用

run_workflow_submit

后，每隔约5秒调用一次

run_workflow_get_status

，并设置

includeOutputs: true

。

进度报告：使用响应中的

elapsedSeconds

，格式为

运行中 — 已完成N/M（X%）· 已耗时Ns

任务完成：逐列出每个输出结果的完整预签名链接，切勿截断
任务失败：调用
```
inspect_run
```
，总结失败原因后终止流程——请勿自动重试

Known featured workflows

已知精选工作流

Slug / keyword	Use case
`product-banners-at-scale`	Product banners from a CSV (data merge), one template + constant image, per-row copy
`retargeting`	Retargeting ads
`localization`	Localization
`generic-banner-advertising`	Generic banners
`contextual-or-native-advertising`	Native ads
`catalogs-and-circulars`	Catalogs
`promotional-pricing-or-discounting`	Pricing promos
`packaging-variants-2d`	Packaging
`standardized-packshots-2d` / `3d`	Packshots
`refreshed-product-imagery-2d`	Product imagery refresh
`composite-imagery-3d`	3D composite
`3d-spin-video`	360 spin video

别名/关键词	使用场景
`product-banners-at-scale`	基于CSV生成批量产品横幅（数据合并），一个模板+固定图片，每行对应不同文案
`retargeting`	重定向广告制作
`localization`	本地化处理
`generic-banner-advertising`	通用横幅广告制作
`contextual-or-native-advertising`	原生广告制作
`catalogs-and-circulars`	商品目录制作
`promotional-pricing-or-discounting`	促销定价活动物料制作
`packaging-variants-2d`	包装设计变体制作
`standardized-packshots-2d` / `3d`	标准化产品实拍图制作（2D/3D）
`refreshed-product-imagery-2d`	产品图片更新（2D）
`composite-imagery-3d`	3D合成图片制作
`3d-spin-video`	360°旋转视频制作

Anti-patterns

反模式

Calling
```
compose_workflow
```
when featured workflow exists and tags match
Sending full graph JSON when
```
session_id
```
is available
Repeating
```
fonts
```
on every merge-data input — use
```
defaults.fonts
```
Sending all input entries when
```
useSessionDefaults: true
```
suffices
Separate submit per image
Skipping user confirmation on missing merge-data text fields

当精选工作流存在且标签匹配时，仍调用
```
compose_workflow
```
当
```
session_id
```
可用时，发送完整的工作流图JSON
在每个数据合并输入项中重复填写
```
fonts
```
——应使用
```
defaults.fonts
```
当
```
useSessionDefaults: true
```
已满足需求时，仍发送所有输入条目
每张图片单独提交一次
缺少数据合并文本字段时，跳过用户确认步骤

References

参考资料

Full run-workflow skill (asset input, publish/save, alert RCA): run-workflow/SKILL.md

完整run-workflow skill（资产输入、发布/保存、故障告警排查）：run-workflow/SKILL.md