creatify-mcp
🎬 创建 MCP 服务器
用于 AI 视频生成的终极 MCP 服务器- 将Creatify AI强大的视频创作功能带给 MCP 生态系统中的每个 AI 助手。
🌟 概述
Creatify MCP 服务器是一款功能全面的模型上下文协议 (MCP) 服务器,它将 Creatify AI 视频生成平台的全部功能开放给 AI 助手、聊天机器人和自动化工具。该服务器基于强大的@tsavo/creatify-api-ts TypeScript 客户端库构建,可将复杂的视频创建工作流程转化为简单的自然语言交互。
🎨高级 MCP 功能:
📝 Prompts - 可重复使用的视频创建模板和工作流程
📊 日志记录- 具有多个严重性级别的结构化日志记录
🔍 进度跟踪- 视频生成过程中的实时更新
🤖 AI 自助-
how_to_useAI 助手理解参数的工具📊 通知- 实时状态更新和进度通知
🎯 这项技术能带来什么
想象一下,你告诉 Claude Desktop: “创建一个 16:9 的 Anna 头像视频,视频中安娜说着‘欢迎观看我们的产品演示’,然后等待它完成。” —— 而这一切真的发生了。这就是 MCP 服务器的强大之处。
🏗️ 内置
Creatify AI API - 全球领先的AI视频生成平台
@tsavo/creatify-api-ts - 综合的 TypeScript 客户端库
模型上下文协议- 标准化人工智能助手集成
TypeScript - 完全类型安全和出色的开发人员体验
Related MCP server: Grok Imagine Video MCP Server
特征
🛠️ MCP 工具(12 个强大的操作)
create_avatar_video- 创建具有唇形同步的 AI 头像视频create_url_to_video- 将网站转换为专业视频generate_text_to_speech- 从文本生成自然语音create_multi_avatar_conversation- 创建包含多个头像对话的视频create_custom_template_video- 使用自定义模板生成视频create_ai_edited_video- 自动编辑和增强视频create_ai_shorts- 制作短视频(适用于 TikTok、Instagram Reels)generate_ai_script- 为视频生成 AI 脚本create_custom_avatar- 设计和创建您自己的自定义头像(DYOA)manage_music- 上传、管理和使用背景音乐create_advanced_lipsync- 具有情绪和手势控制的高级唇形同步how_to_use- 获取任何工具的详细使用信息get_video_status- 检查视频生成任务的状态
📚 MCP 资源(6 个数据源)
creatify://avatars- 可用 AI 头像列表creatify://voices- 可用于文本转语音的声音列表creatify://templates- 可用的自定义视频模板creatify://music- 可用的背景音乐库creatify://credits- 剩余 API 积分creatify://avatar/{avatarId}- 特定头像的详细信息
🏆为什么选择 Creatify MCP 服务器?
🚀完整的 API 覆盖
✅ **12 个 MCP 工具,**涵盖 100% 的 Creatify API 功能
✅ 6 个 MCP 资源,提供全面的数据访问
✅ 5 个常见视频创作场景的工作流程提示
✅**企业级日志记录,**具有 8 个严重级别
🤖 AI 优先设计
✅ 使用
how_to_use工具为 AI 助手提供自我文档✅智能参数验证和错误处理
✅ 视频生成过程中实时进度更新
✅ 具有自动发布的语义版本控制
🎨高级功能
✅ 高级唇形同步中的情绪和手势控制
✅自定义头像创建(DYOA - 设计您自己的头像)
✅ 适用于任何视频类型的AI 脚本生成
✅ TikTok/Instagram/YouTube 的短视频优化
✅背景音乐管理和集成
💼生产就绪
✅ TypeScript提供类型安全性和更好的 IDE 支持
✅ 具有详细上下文的全面错误处理
✅ 用于监控和调试的专业日志记录
✅自动化测试和持续集成
✅语义版本控制,实现可靠更新
📝 MCP 提示(5 个工作流模板)
create-product-demo- 专业产品演示工作流程create-social-content- 参与社交媒体内容创作create-educational-video- 教育和教程视频工作流程create-marketing-campaign- 营销活动视频制作analyze-video-performance- 视频性能分析和优化
先决条件
Node.js 18 或更高版本
Createfy API 凭证(专业版或更高版本)
从Createfy 帐户设置中获取您的 API 凭据
安装
来自 npm(推荐)
npm install -g @tsavo/creatify-mcp从源头
git clone https://github.com/TSavo/creatify-mcp.git
cd creatify-mcp
npm install
npm run build
npm link🎬现场演示示例
创建AI头像视频
User: "Create a professional avatar video with Anna saying 'Welcome to our company!' in 16:9 format"
Claude: I'll create a professional avatar video for you using the Creatify MCP Server.
[Uses create_avatar_video tool with:
- text: "Welcome to our company!"
- avatarId: "anna_costume1_cameraA"
- aspectRatio: "16:9"
- voiceId: "en-US-AriaNeural"]
Your professional avatar video has been created! Anna delivers your message with natural lip-sync and professional presentation.生成TikTok风格的短视频
User: "Make a 30-second TikTok video about coffee brewing tips"
Claude: I'll create an engaging short-form video perfect for TikTok!
[Uses create_ai_shorts tool with:
- prompt: "Quick coffee brewing tips with energetic presentation"
- aspectRatio: "9:16"
- duration: 30
- style: "energetic"]
Your TikTok-ready short video is being generated with optimal 9:16 aspect ratio and engaging content!将网站转换为视频
User: "Turn my product landing page into a marketing video"
Claude: I'll convert your website into a compelling marketing video.
[Uses create_url_to_video tool with:
- url: "https://yourproduct.com"
- aspectRatio: "16:9"
- style: "marketing"]
Your landing page has been transformed into a professional marketing video showcasing your product!人工智能脚本生成
User: "Generate a script for a 60-second product demo video"
Claude: I'll create a compelling script for your product demo.
[Uses generate_ai_script tool with:
- topic: "Product demonstration video"
- scriptType: "commercial"
- duration: 60
- tone: "professional"]
Your script is ready! It includes engaging hooks, clear value propositions, and a strong call-to-action optimized for 60-second format.配置
将您的 Createfy API 凭据设置为环境变量:
export CREATIFY_API_ID="your-api-id"
export CREATIFY_API_KEY="your-api-key"或者创建一个.env文件:
CREATIFY_API_ID=your-api-id
CREATIFY_API_KEY=your-api-key用法
使用 Claude Desktop
添加到您的 Claude Desktop 配置(在 macOS 上为~/Library/Application Support/Claude/claude_desktop_config.json ):
{
"mcpServers": {
"creatify": {
"command": "creatify-mcp",
"env": {
"CREATIFY_API_ID": "your-api-id",
"CREATIFY_API_KEY": "your-api-key"
}
}
}
}🎨高级 MCP 功能
📝使用提示(工作流模板)
AI 助手现在可以使用预定义的工作流模板来处理常见的视频创作场景:
示例:产品演示工作流程
User: "Use the create-product-demo prompt for 'Amazing Widget' with features 'fast, reliable, easy to use' targeting small business owners"
Claude: I'll use the product demo workflow template to create a professional demonstration video.
[Claude automatically follows the complete workflow:
1. Generates an engaging script using generate_ai_script
2. Creates avatar video using create_avatar_video
3. Optimizes for the target audience
4. Includes clear call-to-action]可用的提示模板:
create-product-demo- 专业产品演示create-social-content- TikTok/Instagram/YouTube 内容create-educational-video- 教程和教育内容create-marketing-campaign- 营销和宣传视频analyze-video-performance- 视频优化和分析
📊实时记录和进度
服务器提供具有多个严重程度级别的结构化日志记录:
[INFO] Creatify MCP Server initialized
[INFO] Creating avatar video {avatarId: "anna_costume1_cameraA", aspectRatio: "16:9"}
[INFO] Waiting for avatar video completion...
[INFO] Avatar video completed {videoId: "video_abc123"}日志级别: debug 、 info 、 notice 、 warning 、 error 、 critical 、 alert 、 emergency
🤖 AI自助系统
AI 助手现在可以使用how_to_use工具更好地理解工具参数:
Claude: Let me check how to use the avatar video tool...
[Calls how_to_use tool with toolName: "create_avatar_video"]
[Gets comprehensive documentation with:
- Required parameters with descriptions
- Optional parameters with usage notes
- Real code examples
- Tips and best practices]
Now I understand exactly how to create your avatar video!使用自定义 MCP 客户端
import { Client } from "@modelcontextprotocol/sdk/client/index.js";
import { StdioClientTransport } from "@modelcontextprotocol/sdk/client/stdio.js";
const transport = new StdioClientTransport({
command: "creatify-mcp",
env: {
CREATIFY_API_ID: "your-api-id",
CREATIFY_API_KEY: "your-api-key"
}
});
const client = new Client({
name: "my-client",
version: "1.0.0"
});
await client.connect(transport);
// List available tools
const tools = await client.listTools();
console.log("Available tools:", tools.tools.map(t => t.name));
// Create an avatar video
const result = await client.callTool({
name: "create_avatar_video",
arguments: {
text: "Hello, world! This is an AI-generated video.",
avatarId: "anna_costume1_cameraA",
aspectRatio: "16:9",
waitForCompletion: true
}
});独立服务器
# Set environment variables
export CREATIFY_API_ID="your-api-id"
export CREATIFY_API_KEY="your-api-key"
# Run the server
creatify-mcp人工智能助手的示例提示
一旦配置了 Claude Desktop 或其他 MCP 客户端,您就可以使用自然语言提示,例如:
“创建一个 16:9 的安娜头像视频,视频中安娜说‘欢迎参加我们的产品演示’,然后等待它完成”
“将网站https://example.com转换为宣传视频”
“使用专业声音为‘Hello world’生成文本转语音音频”
“显示所有可用的头像及其详细信息”
“查看我剩余的 Creatify 积分”
“创建两个角色之间的对话来讨论我们的新产品”
API 参考
工具
create_avatar_video
创建具有唇形同步语音的 AI 头像视频。
参数:
text(字符串,必需) - 要朗读的文本avatarId(字符串,必需) - 要使用的头像 IDaspectRatio(“16:9”|“9:16”|“1:1”,必需) - 视频宽高比voiceId(字符串,可选) - 头像的语音 IDwaitForCompletion(boolean,可选) - 等待视频完成
create_url_to_video
将网站 URL 转换为专业视频。
参数:
url(字符串,必需) - 要转换的 URLvisualStyle(字符串,可选) - 视觉样式模板scriptStyle(字符串,可选) - 脚本编写风格aspectRatio(“16:9”|“9:16”|“1:1”,可选) - 视频宽高比waitForCompletion(布尔值,可选)-等待视频完成
generate_text_to_speech
从文本生成自然的语音。
参数:
text(字符串,必需) - 要转换为语音的文本voiceId(字符串,必需) - 要使用的语音 IDwaitForCompletion(布尔值,可选)-等待音频完成
get_video_status
检查视频生成任务的状态。
参数:
videoId(字符串,必需) - 要检查的视频/任务 IDvideoType(字符串,必需) - 任务类型(“lipsync”、“url-to-video”等)
资源
creatify://avatars
返回所有可用 AI 头像及其 ID、名称和元数据的 JSON 列表。
creatify://voices
返回可用于文本转语音生成的所有可用声音的 JSON 列表。
creatify://templates
返回可用自定义视频模板的 JSON 列表。
creatify://credits
返回当前帐户信用余额和使用情况信息。
发展
# Install dependencies
npm install
# Build the project
npm run build
# Run in development mode with auto-reload
npm run dev
# Run tests
npm test
# Lint and format code
npm run check贡献
分叉存储库
创建功能分支(
git checkout -b feature/amazing-feature)提交您的更改(
git commit -m 'Add amazing feature')推送到分支(
git push origin feature/amazing-feature)打开拉取请求
执照
MIT 许可证 - 详情请参阅LICENSE文件。
相关项目
@tsavo/creatify-api-ts- Creatify API 的 TypeScript 客户端模型上下文协议- 协议规范
Creatify AI - AI 视频生成平台
📚 全面的文档
🎬 视频教程
即将推出 - 展示真实使用场景的综合视频教程
📖 API 参考
详细API文档请参见:
Creatify API 文档- 官方 Creatify API 文档
@tsavo/creatify-api-ts 文档- TypeScript 客户端库文档
模型上下文协议规范- MCP 协议细节
🔧 高级配置
环境变量
多变的 | 必需的 | 描述 | 例子 |
| ✅ | 您的 Createfy API ID |
|
| ✅ | 您的 Createfy API 密钥 |
|
| ❌ | 日志级别 |
|
Claude Desktop高级配置
{
"mcpServers": {
"creatify": {
"command": "creatify-mcp",
"env": {
"CREATIFY_API_ID": "your-api-id",
"CREATIFY_API_KEY": "your-api-key",
"MCP_LOG_LEVEL": "info"
},
"args": ["--verbose"]
}
}
}🚀 性能优化
批量操作
对于多个视频创作,请考虑使用批处理功能:
// Example: Create multiple videos efficiently
const videos = await Promise.all([
client.callTool({
name: "create_avatar_video",
arguments: { text: "Video 1", avatarId: "anna", aspectRatio: "16:9" }
}),
client.callTool({
name: "create_avatar_video",
arguments: { text: "Video 2", avatarId: "john", aspectRatio: "16:9" }
})
]);缓存策略
头像/语音列表:缓存 1 小时(很少改变)
视频状态:每 5-10 秒轮询一次活动任务
模板:缓存 24 小时
🔐 安全最佳实践
切勿将 API 密钥提交到版本控制
对所有敏感数据使用环境变量
定期轮换 API 密钥
监控 API 使用情况以检测未经授权的访问
对所有 webhook URL使用 HTTPS
🐛 故障排除
常见问题
“未找到 API 凭证”
# Solution: Set environment variables
export CREATIFY_API_ID="your-api-id"
export CREATIFY_API_KEY="your-api-key"“视频创建失败”
检查您的 Createfy 帐户积分
验证头像/语音 ID 是否存在
确保文本不为空
检查纵横比是否有效
“MCP 连接失败”
验证服务器是否正在运行
检查 Claude Desktop 配置
确保 Node.js 版本 >= 18
调试模式
# Run with debug logging
MCP_LOG_LEVEL=debug creatify-mcp📊 监控与分析
使用情况追踪
监控您的 Createfy API 使用情况:
// Check remaining credits
const credits = await client.readResource({ uri: "creatify://credits" });
console.log(`Remaining credits: ${JSON.parse(credits.contents[0].text).remaining_credits}`);绩效指标
视频制作时间:通常为 2-5 分钟
API 响应时间:通常<2秒
成功率:监控失败的请求
🤝 贡献
欢迎大家投稿!以下是参与方式:
🛠️ 开发设置
# Clone the repository
git clone https://github.com/TSavo/creatify-mcp.git
cd creatify-mcp
# Install dependencies
npm install
# Set up environment variables
cp .env.example .env
# Edit .env with your API credentials
# Run tests
npm test
# Build the project
npm run build
# Run in development mode
npm run dev🧪 测试
# Run all tests
npm test
# Run tests in watch mode
npm run test:watch
# Run type checking
npm run type-check
# Run linting
npm run lint📝 代码风格
我们使用:
ESLint用于代码检查
Prettier用于代码格式化
TypeScript类型安全
提交消息的常规提交
🔄 拉取请求流程
分叉存储库
创建功能分支(
git checkout -b feature/amazing-feature)进行更改
添加新功能测试
确保所有测试通过(
npm test)运行 linting(
npm run lint:fix)提交您的更改(
git commit -m 'feat: add amazing feature')推送到分支(
git push origin feature/amazing-feature)打开拉取请求
📄 许可证
该项目根据 MIT 许可证获得许可 - 有关详细信息,请参阅LICENSE文件。
🙏 致谢
Creatify AI - 提供出色的 AI 视频生成平台
@tsavo/creatify-api-ts - 为该服务器提供支持的强大的 TypeScript 客户端库
Anthropic - 适用于 Claude 和模型上下文协议
MCP 社区- 为实现这种集成的标准化协议
📞 支持
📖 Created API 文档- 官方 API 文档
🐛**报告问题**- 错误报告和功能请求
💬 MCP 社区- 社区讨论
📧**联系作者**- 直接支持
由T Savo用❤️创作
🌐**地平线城市**——开启人工智能革命,加速人类灭绝
让每个开发人员和人工智能助手都能生成人工智能视频——距离人类淘汰更近一步
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
No tool schema history has been recorded yet.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
MCP server for Google Veo AI video generation
MCP server for Hailuo (MiniMax) AI video generation
MCP server for OpenAI Sora AI video generation
MCP server for Luma Dream Machine AI video generation
Related MCP Servers
- AlicenseBqualityDmaintenanceThis MCP server enables any MCP client to generate avatars and videos using the HeyGen API, with tools for managing credits, voices, and avatar video generation.7MIT
- AlicenseAqualityBmaintenanceMCP server for generating, editing, and batch processing videos using xAI's Grok Imagine Video API, with support for text-to-video, image-to-video, and video editing via natural language prompts.4831MIT
- AlicenseNot gradedqualityDmaintenanceMCP server integrating Higgsfield AI API for image generation, video animation, and speech-to-video synthesis with character management and job tracking.4MIT
- AlicenseAqualityCmaintenanceMCP server for faceless YouTube video automation, turning LLM-authored scripts into narrated, captioned videos with AI-generated images and motion.9MIT
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/TSavo/creatify-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server