INDmoney MCP

Name: INDmoney MCP
Author: VIckys-AI-Stuffs

@VIckys-AI-Stuffs

关于 INDmoney MCP

Give Claude AI read-only access to your entire INDmoney portfolio. Fetch Indian stocks, mutual funds, US stocks, gold, credit cards, loans, savings, and Experian credit score — all from Claude Desktop. Uses Playwright browser automation with OTP login and AES-256-GCM encrypted pe

#finance #india #portfolio #stocks #mutual-funds #indmoney #playwright #claude

基本信息

分类

浏览器自动化

传输方式

stdio

发布者

VIckys-AI-Stuffs

提交者

Vickyms Muthu

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "indmoney": {
      "command": "python",
      "args": [
        "-m",
        "mcp_server.main"
      ],
      "env": {
        "PYTHONUNBUFFERED": "1",
        "SESSION_ENCRYPTION_KEY": "<your-32-byte-hex-key>"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is INDmoney MCP?

INDmoney MCP is a Model Context Protocol server that gives Claude read-only access to your INDmoney portfolio — stocks, mutual funds, US stocks, gold, credit cards, liquid savings, loans, and debt. Built with Python + Playwright, it uses browser automation for OTP login and network interception to capture live API data, with encrypted sessions that persist across restarts.

How to use INDmoney MCP?

After installing Python 3.11+, cloning the repository, creating a venv, and installing dependencies (including Playwright Chromium), generate a 32‑byte hex encryption key and configure Claude Desktop’s claude_desktop_config.json with the absolute path to the virtual environment’s Python interpreter, the server script, and the SESSION_ENCRYPTION_KEY in the env block. Restart Claude Desktop, then ask to connect to INDmoney; Claude will open a browser for OTP login, after which you can query portfolio data using any of the 14 MCP tools.

Key features of INDmoney MCP

14 MCP tools for status, connect, disconnect, and portfolio data.
AES-256-GCM encrypted persistent sessions with 12-hour TTL.
Two-level cache (in-memory 5 min + disk 60 min) for instant subsequent queries.
Reliable fetching via direct JWT Bearer calls and page-nav interception fallback.
Read-only — no order placement or fund transfers.
Type-safe with Pydantic models; all responses are structured JSON.

Use cases of INDmoney MCP

Ask Claude for your stock holdings with current P&L.
Query mutual fund performance including XIRR.
Check your US stocks portfolio value in INR.
Get a full portfolio summary with asset allocation and total invested vs current value.
View outstanding credit card balances, due dates, and credit score.

FAQ from INDmoney MCP

What data can I access through this server?

You can access your Indian stock holdings, mutual funds, US stocks, gold, liquid savings/cash, credit cards, loans, debt instruments, and Experian credit score — all read-only.

Do I need to keep any credentials in the config?

No. OTP is entered manually in the browser window opened by Playwright; the server only extracts the JWT from cookies. No credentials are stored in the configuration.

What runtime dependencies are required?

Python 3.11+, Playwright with Chromium, and a MCP‑compatible client (e.g., Claude Desktop). Works on macOS, Linux, and Windows (WSL2).

How are sessions persisted and encrypted?

Sessions are saved to disk encrypted with AES-256-GCM. The encryption key is provided via the SESSION_ENCRYPTION_KEY environment variable. Sessions have a configurable TTL (default 12 hours) and survive server restarts.

Can the server place trades or modify my account?

No. The server is explicitly read-only — there are no tools for placing orders, transferring funds, or modifying your INDmoney account.

浏览器自动化分类下的更多 MCP 服务器

Browser Control MCP

eyalzh

MCP server paired with a browser extension that enables AI agents to control the user's browser.

299

playwright-server MCP server

blackwhite084

189

Fastcrw

Fast, lightweight Firecrawl/Tavily alternative in Rust. Web scraper, crawler & search API with MCP server for AI agents. Drop-in Firecrawl-compatible API (/scrape, /crawl, /search). 2.3x faster than Tavily, 1.5x faster than Firecrawl in 1K-URL benchmarks. 6 MB RAM, single binary.

286