← Back to Skills Marketplace
kylinr

Hailuo Video Generator

by Rong · GitHub ↗ · v1.0.0
cross-platform ✓ Security Clean
759
Downloads
1
Stars
2
Active Installs
1
Versions
Install in OpenClaw
/install hailuo-video-generator
Description
MiniMax Hailuo 视频生成技能 - 使用 S2V-01 模型进行主体参考视频生成。可以生成视频、查询任务状态、下载视频文件。
README (SKILL.md)

MiniMax Hailuo Video Skill

使用 MiniMax Hailuo S2V-01 模型进行主体参考视频生成。

环境配置

API Key 从以下优先级获取:

  1. 环境变量 MINIMAX_API_KEY
  2. 配置文件 ~/.openclaw/openclaw.json 中的 skills.hailuoVideo.apiKey

如需手动设置,在 ~/.openclaw/openclaw.json 中添加:

{
  "skills": {
    "hailuoVideo": {
      "apiKey": "your-api-key-here"
    }
  }
}

API 端点

操作 方法 端点
生成视频 POST https://api.minimax.chat/v1/video_generation
查询状态 GET https://api.minimax.chat/v1/query/video_generation?task_id=xxx
下载视频 GET https://api.minimax.chat/v1/files/retrieve?file_id=xxx

功能

1. 生成视频 (S2V-01 主体参考)

# 使用主体参考图生成视频
curl -s "https://api.minimax.chat/v1/video_generation" \
  -H "Authorization: Bearer $MINIMAX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "S2V-01",
    "prompt": "人物描述动作",
    "subject_reference": [
      {
        "type": "character",
        "image": ["https://example.com/character.jpg"]
      }
    ]
  }'

参数说明:

参数 必填 说明
model 使用 S2V-01
prompt 视频描述,最大2000字符
subject_reference 主体参考图数组
subject_reference[].type 固定为 character
subject_reference[].image 图片URL数组(目前仅支持单张)
prompt_optimizer 是否自动优化prompt,默认true
aigc_watermark 是否添加水印,默认false
callback_url 回调通知地址

图片要求:

  • 格式:JPG, JPEG, PNG, WebP
  • 大小:小于 20MB
  • 尺寸:短边 > 300px,长宽比在 2:5 和 5:2 之间

2. 查询任务状态

curl -s "https://api.minimax.chat/v1/query/video_generation?task_id=YOUR_TASK_ID" \
  -H "Authorization: Bearer $MINIMAX_API_KEY"

状态返回值:

状态 说明
Preparing 准备中
Queueing 队列中
Processing 生成中
Success 成功
Fail 失败

成功响应示例:

{
  "task_id": "176843862716480",
  "status": "Success",
  "file_id": "176844028768320",
  "video_width": 1920,
  "video_height": 1080,
  "base_resp": {"status_code": 0, "status_msg": "success"}
}

3. 下载视频

获取到 file_id 后,调用文件获取接口:

curl -s "https://api.minimax.chat/v1/files/retrieve?file_id=YOUR_FILE_ID" \
  -H "Authorization: Bearer $MINIMAX_API_KEY"

错误码

错误码 说明
0 成功
1002 触发限流
1004 API Key 错误
1008 余额不足
1026 输入内容涉及敏感
1027 生成内容涉及敏感
2013 参数错误
2049 无效 API Key

使用示例

生成皮皮虾视频

# 1. 生成视频
RESPONSE=$(curl -s "https://api.minimax.chat/v1/video_generation" \
  -H "Authorization: Bearer $MINIMAX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "S2V-01",
    "prompt": "一只可爱的小龙虾机器人双手举起做惊讶状,橙色大钳子微微发抖,机械尾巴左右摆动,踉跄着向前冲去",
    "subject_reference": [
      {
        "type": "character",
        "image": ["https://example.com/pipipixia.jpg"]
      }
    ]
  }')

TASK_ID=$(echo $RESPONSE | jq -r '.task_id')
echo "Task ID: $TASK_ID"

# 2. 等待一段时间后查询状态
sleep 60
STATUS=$(curl -s "https://api.minimax.chat/v1/query/video_generation?task_id=$TASK_ID" \
  -H "Authorization: Bearer $MINIMAX_API_KEY" | jq -r '.status')

echo "Status: $STATUS"

# 3. 如果成功,获取文件ID并下载
if [ "$STATUS" = "Success" ]; then
  FILE_ID=$(curl -s "https://api.minimax.chat/v1/query/video_generation?task_id=$TASK_ID" \
    -H "Authorization: Bearer $MINIMAX_API_KEY" | jq -r '.file_id')
  
  # 下载视频
  curl -s "https://api.minimax.chat/v1/files/retrieve?file_id=$FILE_ID" \
    -H "Authorization: Bearer $MINIMAX_API_KEY" \
    -o video.mp4
fi

注意事项

  1. 视频生成通常需要 1-3 分钟
  2. 建议使用轮询查询任务状态,间隔 30-60 秒
  3. 主体参考图必须包含清晰的人物面部
  4. Prompt 应描述期望的动作和场景
Usage Guidance
This skill appears coherent and limited to calling MiniMax video APIs. Before installing: (1) Confirm you trust the MiniMax service at api.minimax.chat and only provide an API key you are willing to use with that provider. (2) If you use the optional callback_url, do not set it to internal or sensitive endpoints unless you understand that task metadata (and possibly links) will be posted there. (3) The examples use jq to parse JSON — install jq if you want to run the examples as-is. (4) Avoid placing unrelated secrets in ~/.openclaw/openclaw.json; the skill only expects skills.hailuoVideo.apiKey.
Capability Analysis
Type: OpenClaw Skill Name: hailuo-video-generator Version: 1.0.0 The skill bundle is a standard integration for the MiniMax Hailuo video generation service. It provides instructions and curl commands to interact with legitimate API endpoints at api.minimax.chat using a user-provided MINIMAX_API_KEY. No evidence of malicious intent, data exfiltration, or unauthorized execution was found in SKILL.md or _meta.json.
Capability Assessment
Purpose & Capability
Name, description, declared env var (MINIMAX_API_KEY), required binary (curl), and the endpoints in SKILL.md all align with a MiniMax video-generation integration. The config path (~/.openclaw/openclaw.json) is used only as an alternate place to read the API key, which is consistent with the stated purpose.
Instruction Scope
SKILL.md contains only curl calls to api.minimax.chat endpoints for create/query/download and instructions to read MINIMAX_API_KEY from env or a specific key in ~/.openclaw/openclaw.json. Minor inconsistency: example usage uses jq to parse responses (e.g., jq -r), but jq is not listed in required binaries. The docs include an optional callback_url parameter (legitimate for asynchronous workflows) — users should be cautious if supplying internal endpoints as callbacks because callbacks send task metadata externally.
Install Mechanism
Instruction-only skill with no install spec and no code files; nothing is downloaded or written to disk by the installer. This is the lowest-risk install profile.
Credentials
Only one credential (MINIMAX_API_KEY) is required and is used by the documented API calls. The single config path is only used to look up the same key. No unrelated secrets or excessive environment access are requested.
Persistence & Privilege
always is false, user-invocable is true, and the skill does not request persistent or elevated platform-wide privileges. It does not modify other skills' configuration.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install hailuo-video-generator
  3. After installation, invoke the skill by name or use /hailuo-video-generator
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of hailuo-video-generator - Generate videos using the MiniMax Hailuo S2V-01 model with subject reference images. - Supports querying video generation task status and downloading generated video files. - Provides API integration with clear environment configuration, including API key setup and usage instructions. - Includes cURL usage examples and guidance on image requirements, error codes, and best practices for optimal results.
Metadata
Slug hailuo-video-generator
Version 1.0.0
License
All-time Installs 2
Active Installs 2
Total Versions 1
Frequently Asked Questions

What is Hailuo Video Generator?

MiniMax Hailuo 视频生成技能 - 使用 S2V-01 模型进行主体参考视频生成。可以生成视频、查询任务状态、下载视频文件。 It is an AI Agent Skill for Claude Code / OpenClaw, with 759 downloads so far.

How do I install Hailuo Video Generator?

Run "/install hailuo-video-generator" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Hailuo Video Generator free?

Yes, Hailuo Video Generator is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Hailuo Video Generator support?

Hailuo Video Generator is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Hailuo Video Generator?

It is built and maintained by Rong (@kylinr); the current version is v1.0.0.

💬 Comments