Orbixio Studio
MCP server for generating and managing YouTube video scripts with human-in-the-loop approval.
README
Orbixio Studio
MCP-native video script generation platform. Generate structured YouTube scripts through Claude Desktop (or any MCP client), review them, and approve before any downstream work begins.
Built as a TypeScript monorepo. No microservices, no workflow engines — just functions.
What it does today (v0)
- 6 MCP tools exposed over stdio:
list_channels,get_channel,generate_script,review_script,approve_script,reject_script - Structured script output validated by zod: hooks, 5 tool entries with honest caveats, CTAs, YouTube metadata
- Human-in-the-loop: scripts must be explicitly approved before any video work begins
- Multi-tenant ready: every row is scoped to a userId (single-user via env var for now)
Prerequisites
- Node.js 20+
- pnpm 9+
- PostgreSQL (local instance or Neon)
- Anthropic API key
Quick Start (< 10 minutes)
1. Clone and install
git clone https://github.com/AhmedAliQadit/orbixio-studio.git
cd orbixio-studio
pnpm install
2. Configure environment
cp .env.example .env
Edit .env with your values:
DATABASE_URL="postgresql://user:password@localhost:5432/orbixio_studio"
ANTHROPIC_API_KEY="sk-ant-..."
ORBIXIO_USER_ID="cldev000000000000000000"
3. Set up the database
# Create the database first (if local Postgres)
createdb orbixio_studio
# Run migrations
pnpm db:migrate
# Seed demo data
pnpm db:seed
4. Build all packages
pnpm build
5. Connect Claude Desktop
Add this to your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"orbixio-studio": {
"command": "node",
"args": ["C:/ABSOLUTE/PATH/TO/orbixio-studio/packages/mcp-server/dist/index.js"],
"env": {
"DATABASE_URL": "postgresql://user:password@localhost:5432/orbixio_studio",
"ANTHROPIC_API_KEY": "sk-ant-...",
"ORBIXIO_USER_ID": "cldev000000000000000000"
}
}
}
}
Replace C:/ABSOLUTE/PATH/TO with the actual path to your clone.
6. Test it
Restart Claude Desktop. You should see "orbixio-studio" in the MCP servers list. Try:
"List my channels, then generate a script about writing sales emails for the first channel."
Claude will call list_channels, then generate_script, and return a structured draft script for your review.
Project Structure
packages/
db/ Prisma schema, client, seed
core/ Pipeline logic (LLM provider, script generator, schemas)
mcp-server/ MCP server wrapping core over stdio
apps/
web/ Stubbed Next.js app (not implemented yet)
Scripts
| Command | Description |
|---|---|
pnpm build |
Build all packages |
pnpm test |
Run all unit tests |
pnpm test:smoke |
Run end-to-end smoke test |
pnpm db:migrate |
Run Prisma migrations |
pnpm db:seed |
Seed demo data |
pnpm lint |
Lint with Biome |
Stack
TypeScript, Prisma, PostgreSQL, @modelcontextprotocol/sdk, @anthropic-ai/sdk, zod, pino, vitest, Biome, pnpm workspaces.
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。