mcp-freshbooks
Production-grade MCP server for FreshBooks. 25 tools for invoices, clients, expenses, payments, time tracking, projects, estimates, and financial reports. OAuth2 with automatic token refresh.
README
mcp-freshbooks
Production-grade MCP server for FreshBooks — 25 tools for invoices, clients, expenses, payments, time tracking, projects, estimates, and financial reports.
Features
- 25 tools covering the full FreshBooks accounting workflow
- OAuth2 authentication with automatic token refresh
- Clean output — summarized lists, formatted details
- Production-grade error handling and rate limit awareness
- Zero cost — uses FreshBooks free developer program
Tools
| Category | Tools | Description |
|---|---|---|
| Auth | freshbooks_authenticate, freshbooks_authenticate_with_code, freshbooks_whoami |
OAuth2 flow + identity |
| Invoices | list_invoices, get_invoice, create_invoice, update_invoice, send_invoice, delete_invoice |
Full invoice lifecycle |
| Clients | list_clients, get_client, create_client, update_client |
Client management |
| Expenses | list_expenses, get_expense, create_expense |
Expense tracking |
| Payments | list_payments, create_payment |
Payment recording |
| Time Tracking | list_time_entries, create_time_entry |
Time entry management |
| Projects | list_projects, create_project |
Project management |
| Estimates | list_estimates, create_estimate |
Estimate creation |
| Reports | get_report |
Profit & loss, tax summary, payments collected |
Quick Start
1. Install
pip install mcp-freshbooks
Or from source:
git clone https://github.com/AlexlaGuardia/mcp-freshbooks.git
cd mcp-freshbooks
pip install .
2. Get FreshBooks API Credentials
- Sign up at freshbooks.com/pages/developer-signup
- Create an OAuth app in the developer portal
- Set redirect URI to
https://localhost:8555/callback - Copy your Client ID and Client Secret
3. Configure
export FRESHBOOKS_CLIENT_ID=your_client_id
export FRESHBOOKS_CLIENT_SECRET=your_client_secret
export FRESHBOOKS_REDIRECT_URI=https://localhost:8555/callback
4. Add to Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"freshbooks": {
"command": "mcp-freshbooks",
"env": {
"FRESHBOOKS_CLIENT_ID": "your_client_id",
"FRESHBOOKS_CLIENT_SECRET": "your_client_secret",
"FRESHBOOKS_REDIRECT_URI": "https://localhost:8555/callback"
}
}
}
}
5. Authenticate
Use the freshbooks_authenticate tool on first use. It will give you a URL to open in your browser. After authorizing, tokens are saved to ~/.mcp-freshbooks/tokens.json and auto-refresh.
Usage Examples
List overdue invoices:
Use list_invoices with status "outstanding"
Create and send an invoice:
Create an invoice for client 12345 with a line item for "Web Development" at $1,500, then send it
Check profitability:
Get the profit and loss report for Q1 2026
Track time:
Create a 2-hour time entry for project 789 with note "API integration work"
Architecture
src/mcp_freshbooks/
├── server.py # MCP server with 25 tool definitions
├── client.py # FreshBooks API client (httpx async)
└── auth.py # OAuth2 flow + token persistence
The server uses the MCP Python SDK with FastMCP for clean tool registration. All API calls go through the async client with automatic token refresh.
Requirements
- Python 3.10+
- FreshBooks account (free trial works for development)
- FreshBooks OAuth app credentials
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。