About Ghost Hunting AI - Service
Ghost Hunting AI is an interactive city tour service powered by LLM, RAG, and MCP that generates guided tours with puzzles for each location based on text files. This is the backend service.
Key features
- 01RAG-based knowledge retrieval
- 02LLM-generated tour content
- 03MCP integration
- 04Text data source processing
- 05Location-specific puzzle generation
Use cases
- 01Creating interactive city tour experiences
- 02Generating location-aware tour narratives
- 03Building AI-powered geographic applications
MCP support
The service uses MCP to connect LLM and RAG systems, providing tour generation capabilities to frontend clients.
Details
From the project READMEGhost Hunting AI - Service
We have developed a service using LLM, RAG, and MCP that generates a guided tour with puzzles for each location, based on your own text files.
Frontend
Link to the frontend repository: KAFRMA-Frontend
Team
Usage
python main.py
Setup Requirements
Logs
You need to create a log directory in the root of the project.
MCP
Create a new directory named credentials in the root of the project. Inside this directory, create a .env file containing the following variables: GOOGLE_DRIVE_FOLDER_ID: Specifies the folder where all data passed to the RAG is stored. OPENROUTER_API_KEY: Stores your LLM API key.
Additional Requirements
Ensure you have Python installed (recommended version: 3.8 or higher). Install the required dependencies using: pip install -r requirements.txt Make sure your API keys are valid and have the necessary permissions. Verify that your Google Drive folder contains the correct data before running the service. Ensure these steps are completed to enable the service to function correctly.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.