MCP Portfolio Server
Fetches and normalizes investment portfolio data from eToro and Interactive Brokers (IBKR), enabling AI agents to consolidate, aggregate, and visualize holdings across brokers.
README
MCP Portfolio Server
A lightweight MCP server with standalone functions to fetch and normalize investment portfolio data from eToro and Interactive Brokers (IBKR) that saves time consuming effort of importing and extracting portfolio data in spreadsheets.
Overview
This project provides three reusable Python files:
etoro_portfolio.py— fetches and normalizes eToro portfolio dataibkr_portfolio.py— fetches and normalizes IBKR portfolio dataportfolio_mcp_server.py— runs as an MCP server
The files themselves do not automatically aggregate positions across brokers. Instead, their outputs are designed to be easily consolidated and visualized using any agent such as Claude, Gemini, or ChatGPT.
MCP Server
The portfolio_mcp_server.py exposes the eToro and IBKR portfolio functions as MCP tools. When running, it allows AI agents (such as Claude, Gemini, or ChatGPT with MCP support) to:
- Call
get_etoro_portfolioto retrieve normalized eToro holdings - Call
get_ibkr_portfolioto retrieve normalized IBKR holdings
These tools return structured portfolio data (symbol, quantity, market value, unrealized P&L, asset class) that the agent can then use to:
- Merge and deduplicate positions across brokers
- Calculate portfolio-level aggregates (total market value, cost basis, P&L, % allocation)
- Classify assets and generate allocation breakdowns
- Create consolidated tables and visualizations (e.g., donut charts, bar charts)
Because the server only provides the raw normalized data, any further data consolidation, aggregation, or visualisation is performed by the agent using the tool outputs.
Installation
git clone https://github.com/<username>/mcp-portfolio-server.git
cd mcp-portfolio-server
uv sync
Usage
eToro
from etoro_portfolio import get_etoro_portfolio
portfolio = get_etoro_portfolio()
IBKR
from ibkr_portfolio import get_ibkr_portfolio
portfolio = get_ibkr_portfolio()
MCP Server
mcp dev portfolio_mcp_server.py
Configuration
Store credentials in a .env file (excluded from the repository):
ETORO_PUBLIC_KEY=your_public_key
ETORO_PRIVATE_KEY=your_private_key
IBKR_TOKEN=your_flex_token
QUERY_ID=your_query_id
License
MIT License
Disclaimer
This project is for personal/educational use and is not affiliated with eToro or Interactive Brokers.
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。