S
SkillNav

图生视频

keevx-image-to-video

by baidu-xiling

Use the Keevx API to convert images to videos. Supports multiple models (V/KL), various resolutions (720p/1080p/4K), and audio generation. Use this skill when the user needs to: (1) Convert images to video (2) Generate video with Keevx (3) Create and query image-to-video tasks (4) Batch image-to-video conversion. Keywords: image to video, Keevx, video generation.

3.9k内容与创意未扫描2026年3月23日

安装

claude skill add --url github.com/openclaw/skills/tree/main/skills/baidu-xiling/keevx-image-to-video

文档

Keevx Image-to-Video Skill

Convert images to high-quality videos via the Keevx API.

Prerequisites

Set the environment variable KEEVX_API_KEY, obtained from https://www.keevx.com/main/home. Documentation: https://docs.keevx.com

bash
export KEEVX_API_KEY="your_api_key_here"

API Endpoints

  • Base URL: https://api.keevx.com/v1
  • Upload image: POST /figure-resource/upload/file (Content-Type: multipart/form-data)
  • Create task: POST /image_to_video (Content-Type: application/json)
  • Query status: GET /image_to_video/{task_id}
  • Auth: All endpoints use Authorization: Bearer $KEEVX_API_KEY
  • Source identifier: All endpoints require the source: skill Header

Model Selection

ParameterModel V (General, Recommended)Model KL (Multi-Reference)
prompt✅ Required, max 1000 chars✅ Required
image_urlRequired for single-image mode, max 20MB❌ Not supported
reference_imagesRequired for multi-image mode, max 3 images, each max 20MB✅ Required, max 7 images, each max 20MB
aspect_ratio✅ 16:9 / 9:16✅ 16:9 / 9:16
aspect_resolution720p / 1080p / 4k720p / 1080p
durationSingle-image: 4/6/8s, Multi-image: fixed 8s5 / 10s
generate_audio✅ Supported, default false❌ Not supported
generate_countOptional 1-4Optional 1-4
callback_urlOptionalOptional

Model Selection Guide

  • Choose Model V: When audio is needed, 4K resolution required, single or multi-image input, high quality requirements
  • Choose Model KL: When multiple reference images are available (2-7), multi-angle showcase needed, product demo videos

Model-Specific Request Bodies

Model V

json
{
  "model": "V",
  "prompt": "Video description (required, max 1000 chars)",
  "image_url": "https://... (required for single-image mode)",
  "reference_images": ["https://..."],
  "aspect_ratio": "16:9",
  "aspect_resolution": "720p | 1080p | 4k",
  "duration": 4,
  "generate_audio": true,
  "generate_count": 1
}

Model KL

json
{
  "model": "KL",
  "prompt": "Video description (required)",
  "reference_images": ["https://..."],
  "aspect_ratio": "16:9",
  "aspect_resolution": "720p | 1080p",
  "duration": 5,
  "generate_count": 1
}

Image Input Handling

User-provided images may be URLs or local file paths, handle accordingly:

  • URL (starts with http:// or https://): Use directly as image_url or reference_images
  • Local file path: Upload via the upload endpoint first, then use the returned URL

Upload Local File

bash
curl --location 'https://api.keevx.com/v1/figure-resource/upload/file' \
  --header 'Authorization: Bearer $KEEVX_API_KEY' \
  --header 'source: skill' \
  --form 'file=@"/path/to/local/image.png"'

Response example:

json
{
  "code": 0,
  "success": true,
  "message": { "global": "success" },
  "result": {
    "url": "https://storage.googleapis.com/..../image.png",
    "fileId": "c5a4676a-...",
    "fileName": "image.png"
  }
}

Extract the image URL from result.url for use as image_url or reference_images. For multiple local files, upload each one and collect all URLs.

Quick Examples

Single Image Generation (Model V)

bash
curl -X POST "https://api.keevx.com/v1/image_to_video" \
  -H "Authorization: Bearer $KEEVX_API_KEY" \
  -H "source: skill" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "V",
    "prompt": "Waves crashing on the beach at sunset",
    "image_url": "https://example.com/beach.jpg",
    "aspect_ratio": "16:9",
    "aspect_resolution": "1080p",
    "duration": 6,
    "generate_audio": true
  }'

Multi-Reference Image Generation (Model KL)

bash
curl -X POST "https://api.keevx.com/v1/image_to_video" \
  -H "Authorization: Bearer $KEEVX_API_KEY" \
  -H "source: skill" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "KL",
    "prompt": "City night timelapse, neon lights flickering",
    "reference_images": [
      "https://example.com/city1.jpg",
      "https://example.com/city2.jpg",
      "https://example.com/city3.jpg"
    ],
    "aspect_ratio": "16:9",
    "aspect_resolution": "1080p",
    "duration": 10,
    "generate_count": 2
  }'

Query Task Status

bash
curl -X GET "https://api.keevx.com/v1/image_to_video/i2v-xxxxxxxx" \
  -H "Authorization: Bearer $KEEVX_API_KEY" \
  -H "source: skill"

Response Format

Task Created Successfully

json
{
  "code": 0,
  "msg": "ok",
  "data": {
    "task_ids": ["i2v-d6b6472bcf724d0399e06d1390cb964e"]
  }
}

Task Query Success

json
{
  "code": 0,
  "msg": "ok",
  "data": {
    "task_id": "i2v-d6b6472bcf724d0399e06d1390cb964e",
    "status": "SUCCEEDED",
    "video_url": "https://storage.googleapis.com/.../sample_0.mp4",
    "thumbnail_url": "https://storage.googleapis.com/.../thumbnail.webp",
    "error_message": "",
    "duration": 6,
    "width": 1920,
    "height": 1080
  }
}

Status Values

PENDING (queued) / PROCESSING (in progress) / SUCCEEDED (completed) / FAILED (failed)

Failure Response

json
{
  "code": 100001,
  "msg": "Parameter error: prompt cannot be empty"
}

Callback Notification

Provide callback_url when creating a task. The system will send a POST request to that URL upon task completion:

json
{
  "code": 0,
  "msg": "ok",
  "task_type": "image_to_video",
  "data": {
    "task_id": "i2v-d6b6472bcf724d0399e06d1390cb964e",
    "status": "SUCCEEDED",
    "video_url": "https://storage.googleapis.com/.../sample_0.mp4",
    "thumbnail_url": "https://storage.googleapis.com/.../thumbnail.webp",
    "error_message": ""
  }
}

Polling Strategy

Video generation may take up to 20 minutes. Recommended: 30-second intervals, max 40 retries.

bash
MAX_RETRIES=40
INTERVAL=30

for i in $(seq 1 $MAX_RETRIES); do
  status=$(curl -s -X GET "$API_BASE/image_to_video/$TASK_ID" \
    -H "Authorization: Bearer $KEEVX_API_KEY" \
    -H "source: skill" | jq -r '.data.status')

  if [ "$status" = "SUCCEEDED" ]; then echo "Success"; break
  elif [ "$status" = "FAILED" ]; then echo "Failed"; break; fi

  sleep $INTERVAL
done

Error Codes

HTTP Status CodeDescription
200Success
400Parameter error
401Authentication failed
404Resource not found
413Request body too large
429Rate limit exceeded
500Internal server error
Business Error CodeDescriptionSolution
100001Parameter errorCheck parameter format and required fields
100002Invalid tokenVerify API Key is correct and active
100003Task not foundVerify task_id is correct
100004Image size exceededCompress image to under 20MB
100005Prompt too longShorten prompt to within 1000 characters
100006Image URL inaccessibleEnsure image URL is publicly accessible
100007Unsupported image formatUse common formats such as JPG, PNG

Notes

  • Generated video/thumbnail URLs are retained for 7 days only; download promptly
  • Max image size: 20MB, supported formats: JPG/PNG/WebP
  • Max prompt length: 1000 characters
  • 429 error: retry after 30s; 500 error: retry after 10s; network error: retry immediately, max 3 times
  • Prompt tips: describe subject, scene, action, mood; include visual details (lighting, color, texture)
  • Narration script: to add narration to the video, use Speak "narration content" in the prompt, e.g. Speak "Welcome to our channel"
  • Image tips: use high-resolution, subject-focused, evenly-lit, well-composed images

相关 Skills

内部沟通

by anthropics

Universal
热门

按公司常用模板和语气快速起草内部沟通内容，覆盖 3P 更新、状态报告、领导汇报、项目进展、事故复盘、FAQ 与 newsletter，适合需要统一格式的团队沟通场景。

✎ 按公司偏好的模板快速产出状态汇报、领导更新和 FAQ，既省去反复改稿，也让内部沟通更统一、更专业。

内容与创意
未扫描111.8k

主题工厂

by anthropics

Universal
热门

给幻灯片、文档、报告和 HTML 落地页快速套用专业配色与字体主题，内置 10 套预设风格并支持现场生成新主题，适合统一品牌或内容视觉。

✎ 主题工厂能帮你把幻灯片、文档到落地页快速统一视觉风格，内置 10 套主题，还能按需即时生成新主题。

内容与创意
未扫描111.8k

文档共著

by anthropics

Universal
热门

围绕文档、提案、技术规格、决策记录等写作任务，按上下文收集、结构迭代、读者测试三步协作共创，减少信息遗漏，写出更清晰、经得起他人阅读的内容。

✎ 写文档、方案或技术规格时容易思路散、信息漏，它用结构化共著流程帮你高效传递上下文、反复打磨内容，还能从读者视角做验证。

内容与创意
未扫描111.8k

相关 MCP 服务

Crypto News Aggregator

by nirholas

热门

免费的加密新闻聚合 MCP，汇集 Bitcoin、Ethereum、DeFi、Solana 与 altcoins 资讯源。

内容与创意
130

MediaWiki

by ProfessionalWiki

让 Large Language Model 客户端无缝连接任意 MediaWiki 站点，可创建、更新、搜索页面，并通过 OAuth 2.0 安全管理内容。

内容与创意16 个工具
72

Transloadit Media Processing

by transloadit

借助 86+ 个云端 media processing robots，处理视频、音频、图像和文档。

内容与创意
71

评论