FinSight is a complete, end-to-end implementation of the Model Context Protocol (MCP), designed to serve as an agentic financial research assistant specifically for the Indian stock market. It operates with a distinct MCP Server component, which exposes stock market functionalities through three MCP primitives: Tools, Resources, and Prompts. The MCP Client then connects to this server, discovers available capabilities, and facilitates a conversational chat loop powered by a local Ollama Large Language Model. Built entirely with free and and open-source tools like Ollama and yfinance, FinSight demonstrates how MCP modularizes AI capabilities, making them reusable and composable across various client applications.
Key Features
01Agentic Indian stock market research assistant
02Multi-stock comparison with side-by-side metrics and auto-rankings
030 GitHub stars
04Automated deep analysis reports using parameterised MCP prompt templates
05Real-time stock price and fundamental data fetching via yfinance
06Keyword router for direct tool calls and LLM for complex query interpretation