# 🇮🇳 mftool-mcp > **MCP Server for publicly available Indian Mutual Funds data** > Powered by [mftool](https://github.com/NayakwadiS/mftool) · Works with Claude, Cursor, and any MCP-compatible LLM client [](https://pypi.org/project/mftool-mcp/) [](https://www.python.org/downloads/) [](https://opensource.org/licenses/MIT) [](https://modelcontextprotocol.io) --- ## What is this? `mftool-mcp` is a [Model Context Protocol (MCP)](https://modelcontextprotocol.io) server that gives any LLM client access to **Indian Mutual Fund data**. Ask your AI assistant questions like: - *"What is the current NAV of SBI Bluechip Direct Growth?"* - *"Show me 3-year returns for HDFC Midcap Opportunities Fund"* - *"List all schemes under Mirae Asset AMC"* - *"Find all flexi cap funds and compare their 5-year returns"* --- ## Tools Available | Tool | Description | |---|---| | `get_scheme_quote` | Live NAV for any scheme by AMFI scheme code | | `get_bulk_quotes` | Live NAV for multiple schemes at once (concurrent) | | `get_scheme_historical_nav` | Full NAV history for a scheme (all available dates) | | `get_scheme_historical_nav_for_dates` | NAV history filtered to a specific date range (DD-MM-YYYY) | | `get_scheme_history` | NAV history via BSE/YFinance code with period or date range | | `get_scheme_codes` | All AMFI scheme codes and names | | `get_available_schemes` | All schemes under a specific AMC | | `get_scheme_details` | Fund house, type, category, start date (AMFI code) | | `get_scheme_info` | Rich scheme info — AUM, returns, ratings (BSE/YFinance code) | | `is_valid_scheme_code` | Validate an AMFI numeric scheme code | | `is_valid_new_scheme_code` | Validate a BSE/new format scheme code | | `search_schemes` | Search schemes by name keyword with relevance ranking | | `search_schemes_by_amc` | Search schemes within a specific AMC | | `search_schemes_by_type` | Search schemes by type/category (equity, debt, elss, etc.) | | `get_average_aum` | Average AUM (domestic & overseas) for all AMCs by quarter | | `get_equity_scheme_performance` | Daily performance for all equity funds (1Y/3Y/5Y returns) | | `get_debt_scheme_performance` | Daily performance for all debt funds (1Y/3Y/5Y returns) | | `get_hybrid_scheme_performance` | Daily performance for all hybrid funds (1Y/3Y/5Y returns) | | `get_solution_scheme_performance` | Daily performance for solution-oriented funds (retirement, children) | | `get_other_scheme_performance` | Daily performance for index funds and Fund of Funds | | `calculate_returns` | Calculate SIP returns — absolute return % and annualised IRR | | `get_cache_stats` | View hit/miss stats for NAV and scheme code caches | | `clear_cache` | Clear all cached data to force fresh fetches | | `set_cache_enabled` | Enable or disable mftool caching globally | --- ## Quick Start ### Using `uvx` (Recommended — no install needed) ```bash uvx mftool-mcp ``` ### Using `pip` ```bash pip install mftool-mcp mftool-mcp ``` --- ## Claude Desktop Setup Add this to your `claude_desktop_config.json`: **macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json` **Windows:** `%APPDATA%\Claude\claude_desktop_config.json` ```json { "mcpServers": { "mftool-mcp": { "command": "uvx", "args": ["mftool-mcp"] } } } ``` Restart Claude Desktop. You'll see **mftool-mcp** listed under connected tools. --- ## Cursor / Other MCP Clients In your MCP client settings, add: ```json { "mftool-mcp": { "command": "uvx", "args": ["mftool-mcp"] } } ``` --- ## Example Conversations **Finding a fund:** > You: "Find all SBI midcap mutual fund scheme codes" > Claude: *calls `search_schemes_by_amc(amc_name='sbi', query='midcap')`* > Claude: "Here are the SBI midcap schemes: SBI Magnum Midcap Fund - Direct Growth (code: 125497)..." **Live NAV:** > You: "What's the current NAV of scheme 125497?" > Claude: *calls `get_scheme_quote(scheme_code='125497')`* > Claude: "SBI Magnum Midcap Fund - Direct Growth: NAV ₹234.56 (as of 05-Apr-2025)" **Performance comparison:** > You: "Compare 5-year returns for large cap equity funds" > Claude: *calls `get_equity_scheme_performance()`* > Claude: "Here's the 5-year return comparison for Large Cap funds: ..." --- ## Data Source All data is sourced via the `mftool` library. Data accuracy depends on published figures. This tool is for **informational purposes only** and does not constitute financial advice. --- ## Development ```bash git clone https://github.com/NayakwadiS/mftool-mcp cd mftool-mcp pip install -e ".[dev]" # Run the server locally python -m mftool_mcp.server ``` --- ## Acknowledgements - [mftool](https://github.com/NayakwadiS/mftool) — the underlying data library ## Associated Initiatives - NSE Stock predictions *[NSE-Neuron](https://github.com/NayakwadiS/NSE-Neuron)*