📦

Unifeast Chatbot

Food recommendation chatbot using LangChain, MCP servers, and Pinecone semantic search. ML and AI Part of the project Unifeast.

0 installs

Trust: 34 — Low

Rag

Ask AI about Unifeast Chatbot

Powered by Claude · Grounded in docs

I know everything about Unifeast Chatbot. Ask me about installation, configuration, usage, or troubleshooting.

0/500

Documentation

UniFeast MCP Chatbot

Food recommendation chatbot using Amazon Bedrock Inline Agents and MCP servers.

Prerequisites

AWS account with Bedrock access
AWS CLI installed and configured
Python 3.11+
Docker installed and running
Claude 3.5 Sonnet model access enabled

Quick Start

Install dependencies

pip install -r requirements.txt

Start PostgreSQL database

docker-compose up -d postgresql

Set up AWS credentials

export AWS_PROFILE=your-profile
export AWS_REGION=us-east-1

Test database connection

# Verify PostgreSQL is running
docker-compose ps

Project Structure

mcp-chatbot/
├── src/
│   ├── mcp_servers/     # 4 MCP servers (users, food, crowd, summary)
│   └── inline_agent_sdk.py  # Main SDK
├── config/
│   └── postgresql/      # Database initialization
└── docker-compose.yml   # PostgreSQL only

Next Steps

Create MCP servers (Phase 2)
Create InlineAgent SDK (Phase 2)
Test integration (Phase 3) # Force redeploy - Sun Oct 5 05:39:49 BST 2025

Force complete Railway rebuild - Sun Oct 5 05:46:58 BST 2025