StackA2A
enterpriselanggraphpython

MathTutor

19

by nottpande

An agentic AI assistant, that is helping users with math queries to solve and explain them to the user.

1 starsUpdated 2025-07-01MIT
Quality Score19/100
Community
7
Freshness
17
Official
30
Skills
10
Protocol
30
🔒 Security
20

Getting Started

1Clone the repository
$ git clone https://github.com/nottpande/MathTutor
2Navigate to the project
$ cd MathTutor
3Install dependencies
$ pip install -r requirements.txt
4Run the agent
$ python main.py

README

MathTutor

Math Tutor is an intelligent, agentic AI assistant designed to help users solve math problems using a combination of document retrieval, web search, and reasoning via LLMs. The system is modular, agent-based, and built using modern protocols like Model Context Protocol (MCP) and Agent to Agent (A2A) Protocol, making it extensible and production-ready.


Features

  • Retrieves relevant context from a vector database (AstraDB)
  • Falls back to web search if vector search yields no results
  • Uses LLMs for reasoning, extracting and solving math questions
  • MCP-based microservices for modularity and scalability
  • LangGraph orchestration for flexible agent workflows
  • A2A-compatible responses with structured artifacts
  • Automatic result saving (raw + cleaned formats) for analysis

Capabilities

StreamingPush NotificationsMulti-TurnAuth: none
agentic-aiagentic-workflowlangchainlanggraphmcp
View on GitHub