fireMCP (Braze Write Server)

fireMCP (Braze Write Server)

A write-enabled extension for the official Braze MCP server that allows users to manage profiles, trigger campaigns, and handle catalogs. It includes safety features such as dry run mode and workspace validation to ensure secure operations in test environments.

Category
访问服务器

README

MCP Write Server

A NON-OFFICIAL write-enabled extension of the official Braze MCP server, designed for Solution Engineers to create demos, trigger campaigns, and manage test environments.

⚠️ Important Safety Notice

This server enables WRITE OPERATIONS to your Braze workspace. It includes multiple safety mechanisms:

  • Workspace validation: Only works with demo/POC/test workspaces by default
  • Rate limiting: Prevents accidental spam
  • Dry run mode: Test operations without making changes
  • Confirmation requirements: Destructive operations need explicit confirmation
  • Production blocking: Prevents writes to production workspaces unless explicitly enabled

Features

Write Operations

  • ✅ User profile management (attributes, events, purchases)
  • ✅ Campaign triggering and management
  • ✅ Canvas triggering and management
  • ✅ Catalog item management
  • ✅ Content block management
  • ✅ User deletion and identification

Safety Features

  • ✅ Environment-based workspace validation
  • ✅ Rate limiting per operation type
  • ✅ Dry run mode for all operations
  • ✅ Confirmation requirements for destructive operations
  • ✅ Comprehensive logging and audit trail

Installation

pip install -e .

Or with uv:

uv pip install -e .

Configuration

Environment Variables

# Required
BRAZE_API_KEY=your-api-key-here
BRAZE_BASE_URL=https://rest.iad-01.braze.com

# Safety Configuration
BRAZE_WRITE_ENABLED=true                    # Enable write operations
BRAZE_ALLOW_PRODUCTION=false                # Block production workspaces
BRAZE_ALLOWED_WORKSPACES=demo-,poc-,test-  # Allowed workspace patterns
BRAZE_DRY_RUN_DEFAULT=false                 # Default dry run mode

# Rate Limits
BRAZE_MAX_SENDS_PER_HOUR=1000              # Max campaign/canvas sends per hour
BRAZE_MAX_CATALOG_UPDATES_PER_MIN=100      # Max catalog updates per minute

MCP Client Configuration

Add to your MCP client configuration (e.g., Claude Desktop):

{
  "mcpServers": {
    "braze-write": {
      "command": "python",
      "args": ["-m", "braze_mcp_write.main"],
      "env": {
        "BRAZE_API_KEY": "your-api-key",
        "BRAZE_BASE_URL": "https://rest.iad-01.braze.com",
        "BRAZE_WRITE_ENABLED": "true",
        "BRAZE_ALLOWED_WORKSPACES": "demo-,poc-,test-"
      }
    }
  }
}

Usage Examples

Update User Attributes

await update_user_attributes(
    ctx,
    external_id="user_12345",
    attributes={
        "first_name": "John",
        "last_name": "Doe",
        "email": "[email protected]",
        "subscription_tier": "premium"
    }
)

Track Custom Event

await track_event(
    ctx,
    event_name="completed_tutorial",
    external_id="user_12345",
    properties={
        "tutorial_name": "Getting Started",
        "completion_time_seconds": 180
    }
)

Send Campaign

await send_campaign(
    ctx,
    campaign_id="your-campaign-id",
    recipients=[{
        "external_user_id": "user_12345"
    }]
)

Trigger Canvas

await trigger_canvas(
    ctx,
    canvas_id="your-canvas-id",
    recipients=[{
        "external_user_id": "user_12345"
    }],
    canvas_entry_properties={
        "promo_code": "DEMO2024"
    }
)

Dry Run Mode

Test any operation without making changes:

result = await update_user_attributes(
    ctx,
    external_id="user_12345",
    attributes={"test_field": "test_value"},
    dry_run=True  # No actual API call made
)

Architecture

braze_mcp_write/
├── models/              # Pydantic models for validation
│   ├── errors.py
│   └── responses.py
├── tools/               # MCP tool implementations
│   ├── users_write.py   # User tracking operations
│   ├── campaigns_write.py
│   ├── canvas_write.py
│   ├── catalogs_write.py
│   └── content_blocks_write.py
├── utils/               # Utilities
│   ├── context.py       # Context management
│   ├── http.py          # HTTP client
│   ├── logging.py       # Logging setup
│   └── safety.py        # Safety mechanisms
├── registry_builder.py  # Auto-discovery system
├── server.py            # FastMCP server
└── main.py              # Entry point

Development

Running Tests

pytest

Code Formatting

black .
ruff check .

Safety Best Practices

  1. Always use demo workspaces: Set up dedicated demo workspaces with the "demo-", "poc-", or "test-" prefix in the URL
  2. Test with dry run first: Use dry_run=True to validate operations before execution
  3. Monitor rate limits: Be aware of Braze API rate limits for your workspace
  4. Use confirmation for destructive ops: Operations like delete_user require confirm=True
  5. Review logs: Check logs regularly for unexpected behavior

Differences from Official Braze MCP

This server extends the official read-only Braze MCP server with:

  • Write operations for campaigns, canvas, users, catalogs, and content blocks
  • Comprehensive safety mechanisms to prevent accidental production writes
  • Rate limiting to prevent API abuse
  • Dry run mode for testing
  • Enhanced logging and audit trails

License

MIT License - See LICENSE file for details

Disclaimer

This is a developer tool for creating demos and POCs. Use with caution and always verify your workspace configuration before enabling write operations.

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选