Sync MCP Server
Enables AI agents to create lipsync videos, manage assets, and check generation status via the Sync API.
README
Sync MCP Server
An open-source Model Context Protocol (MCP) server for the Sync API. Gives AI agents the ability to create lipsync videos, manage assets, check generation status, and more.
Tools are auto-generated from the Sync OpenAPI spec at startup. As new API endpoints ship, they become available to agents automatically — no server update needed.
Supported Clients
| Client | Status | Transport |
|---|---|---|
| Claude Web (claude.ai) | Supported | Remote (HTTP + OAuth) |
| Claude Desktop | Supported | Local (stdio) |
| ChatGPT Desktop | Supported | Local (stdio) |
| Claude Code | Supported | Local (stdio) |
| Cursor | Supported | Local (stdio) |
| Windsurf | Supported | Local (stdio) |
| Codex CLI | Supported | Local (stdio) |
| Any MCP-compatible client | Supported | Local (stdio) |
Quick Start
Claude Web (claude.ai)
No installation required — connect directly from your browser.
- Go to claude.ai → Settings → Integrations
- Click Add custom connector
- Enter Name:
Syncand URL:https://mcp.sync.so/mcp - Click Add, then Connect
- Log in with your Sync account when prompted
No API key needed — authentication is handled via OAuth.
Claude Code
claude mcp add sync -- npx -y @sync.so/mcp-server --api-key YOUR_API_KEY
Or add to .mcp.json in your project root:
{
"mcpServers": {
"sync": {
"command": "npx",
"args": ["-y", "@sync.so/mcp-server"],
"env": {
"SYNC_API_KEY": "your-api-key"
}
}
}
}
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"sync": {
"command": "npx",
"args": ["-y", "@sync.so/mcp-server"],
"env": {
"SYNC_API_KEY": "your-api-key"
}
}
}
}
Cursor
Add to .cursor/mcp.json in your project:
{
"mcpServers": {
"sync": {
"command": "npx",
"args": ["-y", "@sync.so/mcp-server"],
"env": {
"SYNC_API_KEY": "your-api-key"
}
}
}
}
Without an API key (interactive login)
Omit SYNC_API_KEY and the server will start a device auth flow on first run:
{
"mcpServers": {
"sync": {
"command": "npx",
"args": ["-y", "@sync.so/mcp-server"]
}
}
}
You'll be prompted to visit a URL and enter a code. After approval, the token is cached at ~/.config/sync/mcp-credentials.json.
Getting an API Key
- Sign up at sync.so
- Go to your dashboard settings
- Generate an API key
See the authentication guide for details.
Available Tools
Tools are dynamically generated from the Sync API. Core tools include:
| Tool | Description |
|---|---|
generate_create-generation |
Create a lipsync video from video + audio inputs |
generate_get-generation |
Get generation status — poll until COMPLETED |
generate_get-generations |
List recent generations |
models_get |
List available lipsync models |
assets_get-all |
List all assets in your organization |
assets_get |
Get a specific asset by ID |
generations_estimate-cost |
Estimate generation cost before creating |
generations_get-by-id |
Get a generation by ID |
Example Prompts
Once configured, ask your AI agent:
- "List available Sync models"
- "Create a lipsync video with this video URL and audio URL using the lipsync-2 model"
- "Check the status of generation gen-abc123"
- "Show me my recent generations"
- "How much would it cost to generate a 30-second video?"
CLI Options
Usage: sync-mcp [options]
Options:
--api-key <key> API key (or set SYNC_API_KEY env var)
--base-url <url> API base URL (default: https://api.sync.so)
--transport <type> stdio (default) or http
--port <port> HTTP port (default: 3002, only with --transport http)
-h, --help Show this help message
Environment Variables
| Variable | Description | Default |
|---|---|---|
SYNC_API_KEY |
Your Sync API key (stdio transport) | — |
SYNC_BASE_URL |
API base URL | https://api.sync.so |
MCP_ISSUER_URL |
OAuth issuer URL (HTTP transport only) | — |
OAUTH_REGISTRATION_SECRET |
Shared secret for client registration (HTTP transport only) | — |
How It Works
- On startup, the server fetches the OpenAPI spec from
{baseUrl}/api-json - Parses all public endpoints into operation definitions
- Converts each operation into an MCP tool with a Zod input schema
- Registers tools on the MCP server
- Each tool call makes an authenticated HTTP request to the Sync API
This means new API endpoints are automatically available — just restart the MCP server.
Programmatic Usage
You can also use the server as a library:
import { createSyncMcpServer, resolveConfig } from '@sync.so/mcp-server';
const config = resolveConfig({ apiKey: 'your-key' });
const server = await createSyncMcpServer(config);
Development
# Install dependencies
npm install
# Build
npm run build
# Run tests
npm test
# Lint
npm run lint
# Type check
npm run typecheck
Local HTTP Transport Testing
To test the HTTP transport (OAuth flow) locally:
- Copy
.env.exampleto.envand fill in the values - Start a tunnel to expose localhost:
ngrok http 3002 - Update
MCP_ISSUER_URLin.envwith the ngrok URL - Start the server:
source .env node dist/cli.js --transport http --base-url $SYNC_BASE_URL --port 3002 - Verify it works:
curl https://<ngrok-url>/health curl https://<ngrok-url>/.well-known/oauth-authorization-server - Add
https://<ngrok-url>/mcpas a custom connector in Claude Web
Learn More
- Sync Documentation — Full API reference and guides
- Sync API Reference — Endpoint documentation
- MCP Protocol — Learn about the Model Context Protocol
- Sync Website — Sign up and get started
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。