Deploy and Run Seedance Video Generation Skill in Feishu via OpenClaw

Overview

The Seedance video‑generation Skill provides text‑to‑video and image‑to‑video capabilities directly from a Feishu chat. It runs on an OpenClaw instance and communicates with a Feishu bot, allowing users to generate short videos by sending natural‑language prompts.

Prerequisites

OpenClaw installed (via Volcano Engine one‑click deployment or manual installation).

Feishu developer account with permission to create a custom bot.

Volcano Engine Coding Plan API key (optional, required for advanced features).

Deploy OpenClaw

Open the Volcano Engine console and select “One‑click Deploy OpenClaw”.

Purchase the deployment package and choose a subscription tier.

In the advanced configuration, enable the Coding Plan API if you have a Coding Plan.

Configure Feishu Bot

Create a custom bot in the Feishu Open Platform and grant the following permission scopes (JSON example):

{
  "scopes": {
    "tenant": [
      "docs:document.content:read",
      "docs:document.media:upload",
      "im:chat:read",
      "im:message:send_as_bot"
    ],
    "user": [
      "contact:contact.base:readonly"
    ]
  }
}

Event & Callback

In the Feishu developer console navigate to Development Configuration → Event & Callback .

Add a subscription for the message.receive event and select “Long connection” as the receiver.

App ID and Secret

Obtain the App ID and App Secret from the bot’s credential page and paste them into OpenClaw’s message‑channel integration settings for Feishu.

Install Seedance Video Skill

Two installation methods are supported:

Chat command (run in a Feishu conversation):

帮我通过clawhub安装一下这个skill：https://clawhub.ai/JackyCSer/seedance-video-generation

Manual installation :

Download the zip package from the same URL.

Extract it to /Users/your_user_name/.openclaw/skills/.

Verify the skill appears at http://127.0.0.1:18789/skills.

Configure API Keys on the Instance

Set the Volcano Engine Coding Plan API key (if applicable) and the ARK API key in the OpenClaw terminal:

export ARK_API_KEY='<your-api-key>'

Generate Videos

After configuration, send prompts in Feishu to generate videos, for example:

“帮我用seedance-video-generation这个skill根据你的记忆生成我的一天”。

“生成一个15 s左右的视频，内容是一首古诗：枯藤老树昏鸦，小桥流水人家”。

Supported Seedance Models

Seedance 1.5 Pro – Model ID doubao-seedance-1-5-pro-251215 – Text‑to‑video, Image‑to‑video (first frame, first + last frame), Audio support, Draft mode.

Seedance 1.0 Pro – Model ID doubao-seedance-1-0-pro-250428 – Text‑to‑video, Image‑to‑video (first frame, first + last frame).

Seedance 1.0 Pro Fast – Model ID doubao-seedance-1-0-pro-fast-250528 – Text‑to‑video, Image‑to‑video (first frame only).

Seedance 1.0 Lite T2V – Model ID doubao-seedance-1-0-lite-t2v-250219 – Text‑to‑video only.

Seedance 1.0 Lite I2V – Model ID doubao-seedance-1-0-lite-i2v-250219 – Image‑to‑video (first frame, first + last frame, reference images 1‑4).

Default Model

The default model is doubao-seedance-1-5-pro-251215, which includes audio support.

Switching Models

To use a different model, specify its Model ID in the conversation with OpenClaw (e.g., “use model doubao-seedance-1-0-lite-t2v-250219 ”).

Future Roadmap

Seedance 2.0 will be available once the Volcano Engine Seedance 2.0 API reaches General Availability. It can already be tested in the Volcano Engine Experience Center.

Optional Installation Script

curl -fsSL https://openclaw.tos-cn-beijing.volces.com/setup.sh | bash -s -- \
    --ark-coding-plan "true" \
    --ark-api-key "<your-api-key>" \
    --ark-model-id "doubao-seedance-1-5-pro-251215" \
    --feishu-app-id "<your-feishu-app-id>" \
    --feishu-app-secret "<your-feishu-app-secret>"