Add context-aware conversational AI assistant for Sunbird Ed (Issue #515)

[techySPHINX]

🧠 Summary

This PR introduces a context-aware conversational AI assistant as outlined in Issue #515. The assistant leverages the Model Context Protocol (MCP) and LangChain's function-calling to enable intelligent interactions with Sunbird Ed APIs. It is designed to be role-aware, extensible, and modular across deployments.

---

🔍 Key Features

• 🌐 Integrated FastAPI backend with CLI or web interface support
• 🔧 LangChain agent with role-based tool access (learner, mentor, admin)
• 🔐 Authentication and authorization layers for API security
• 🧠 Session management for user and installation-specific contexts
• 📦 Support for mocked Sunbird Ed APIs:
  • Course metadata
  • Enrollment status
  • Learning progress
  • Admin operations
• 🧰 Tool schemas modularized for easy API replacement or extension

---

📁 Project Structure Highlights

• app/agent/session_manager.py: Handles user session and context
• app/tool_schemas/: Implements API interaction logic
• app/tool_registry.py: Registers tools dynamically based on role
• app/agent_setup.py: Initializes the LangChain agent

---

🧪 Testing & Simulation

• Sunbird Ed APIs are mocked for development
• Easily replaceable with real endpoints in production environments

---

🧩 Future Scope

• Add more DPG integrations
• Extend toolset for deeper role-based capabilities
• Add persistence for session data

---