Stream
Stream MCP server https://streampay.sa/
README
Stream MCP Server
An MCP server for the Stream (streampay.sa) payment platform, built with FastMCP.
Exposes 27 tools across six resource domains — payment links, customers, products, coupons, invoices, and payments — plus a read-only OpenAPI documentation resource.
Quick Start
1. Install
# Clone & install in editable mode
git clone <repo-url> stream-mcp-server && cd stream-mcp-server
pip install -e ".[dev]"
2. Configure
cp .env.example .env
# Edit .env and set your Stream API key:
# STREAM_API_KEY=sk_live_…
| Variable | Default | Description |
|---|---|---|
STREAM_API_KEY |
(required for stdio mode) | Your Stream API key |
STREAM_BASE_URL |
https://stream-app-service.streampay.sa |
API base URL |
STREAM_TIMEOUT |
30 |
Request timeout (seconds) |
STREAM_MAX_RETRIES |
2 |
Retry count for 429 / 5xx |
HOST |
0.0.0.0 |
Remote server bind host (stream-mcp-remote) |
PORT |
8000 |
Remote server bind port (stream-mcp-remote) |
3. Run
Local stdio mode (recommended)
stream-mcp
Remote HTTP mode (URL clients)
# No STREAM_API_KEY needed on the server process in remote mode
HOST=0.0.0.0 PORT=8000 stream-mcp-remote
Endpoint:
http://localhost:8000/mcp
MCP Client Configuration (Claude Desktop / Cursor / VS Code)
Use one of these two patterns in your MCP config file (claude_desktop_config.json or mcp.json).
Option A: stdio
{
"mcpServers": {
"stream": {
"command": "stream-mcp",
"env": {
"STREAM_API_KEY": "sk_live_your_key_here"
}
}
}
}
Option B: remote URL (stream-mcp-remote)
{
"mcpServers": {
"stream": {
"url": "http://localhost:8000/mcp",
"headers": {
"Authorization": "Bearer sk_live_your_key_here"
}
}
}
}
Available Tools
Payment Links
| Tool | Description |
|---|---|
create_payment_link |
Create a new checkout / payment link |
list_payment_links |
Paginated list with optional status filter |
get_payment_link |
Get a single payment link by ID |
deactivate_payment_link |
Deactivate / archive a payment link |
Customers
| Tool | Description |
|---|---|
create_customer |
Create a customer with name, email, phone, metadata |
list_customers |
Paginated list of customers |
get_customer |
Get a single customer by ID |
update_customer |
Update customer fields |
delete_customer |
Soft-delete a customer |
Products
| Tool | Description |
|---|---|
create_product |
Create a one-time or recurring product |
list_products |
List products with optional type filter |
get_product |
Get a single product by ID |
update_product |
Update product name, description, or price |
archive_product |
Archive a product |
Coupons
| Tool | Description |
|---|---|
create_coupon |
Create a fixed or percentage discount coupon |
list_coupons |
List coupons with optional status filter |
get_coupon |
Get a single coupon by ID |
deactivate_coupon |
Deactivate a coupon |
Invoices
| Tool | Description |
|---|---|
create_invoice |
Create a ZATCA-compliant invoice |
list_invoices |
List invoices with filters |
get_invoice |
Get a single invoice by ID |
send_invoice |
(Re)send an invoice via email / SMS |
void_invoice |
Void / cancel an unpaid invoice |
Payments
| Tool | Description |
|---|---|
list_payments |
List payments with filters |
get_payment |
Get payment details |
refund_payment |
Issue a full or partial refund |
Resources
| Resource URI | Description |
|---|---|
stream://docs/openapi |
Full Stream OpenAPI spec (cached, auto-refreshed) |
Remote Deployment (Hosted URL)
You can deploy the MCP server as a hosted URL so users connect to it remotely.
Each user passes their own Stream API key as a Bearer token.
1. Run locally (remote mode)
# No STREAM_API_KEY needed — each user provides their own
stream-mcp-remote
# → Listening on http://0.0.0.0:8000
# Custom host/port
HOST=0.0.0.0 PORT=3000 stream-mcp-remote
2. Deploy with Docker
docker build -t stream-mcp .
docker run --rm -p 8000:8000 stream-mcp
3. How users connect (remote)
Users add this to their MCP client config:
Claude Desktop / VS Code:
{
"mcpServers": {
"stream": {
"url": "https://your-domain.com/mcp",
"headers": {
"Authorization": "Bearer sk_live_YOUR_STREAM_API_KEY"
}
}
}
}
Each user passes their own Stream API key as the Bearer token. The server never stores keys — they are used only for the duration of the session.
Project Structure
src/stream_mcp/
├── server.py # FastMCP app entry-point (local + remote modes)
├── config.py # Settings from env vars
├── client.py # Async HTTP client (auth, retries, errors)
├── auth.py # Bearer token middleware (remote mode)
├── helpers.py # get_client() — resolves per-request StreamClient
├── models/ # Pydantic v2 request/response models
│ ├── payment_links.py
│ ├── customers.py
│ ├── products.py
│ ├── coupons.py
│ ├── invoices.py
│ └── payments.py
└── tools/ # FastMCP tool definitions
├── __init__.py # Registers all tools
├── payment_links.py
├── customers.py
├── products.py
├── coupons.py
├── invoices.py
├── payments.py
└── docs.py # OpenAPI resource
Adding a new resource domain = add one file in models/, one in tools/, and one import line in tools/__init__.py.
Error Handling
All tools catch StreamAPIError and return a structured dict instead of raising:
{
"error": true,
"code": 422,
"message": "Validation failed: …"
}
This ensures the LLM agent always receives a usable response.
Development
# Install with dev dependencies
pip install -e ".[dev]"
# Run tests
pytest tests/ -v
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。