Rag chatbot with a localhost MCP server
LocalAbout Rag chatbot with a localhost MCP server
A RAG-based HR chatbot that answers workplace-rules questions using uploaded PDFs. It combines document retrieval with a Streamlit chat interface and GPT-4 answer generation.
Key features
- 01Upload and parse PDFs
- 02Split text into chunks
- 03Index documents in memory
- 04Retrieve relevant passages by similarity
- 05Generate answers with GPT-4
- 06Streamlit chat interface
Use cases
- 01Ask questions about rules in uploaded workplace PDFs
- 02Retrieve relevant passages from indexed documents
MCP support
The localhost MCP server coordinates tools for document indexing, retrieval, and answer generation.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMERag chatbot with a localhost MCP server
Building a RAG-based HR chatbot for providing rules in the workplace with the localhost MCP server as a function-calling hub
Overview
This project implements a Retrieval-Augmented Generation (RAG) chatbot using Streamlit and the MCP server. Users can upload PDF files, and the chatbot retrieves relevant information from the PDFs to answer natural language questions. The system leverages OpenAI models, LangChain utilities, and an in-memory vector store for efficient document retrieval.
Features
-
MCP Tool Integration:
Tool orchestration with MCP ensures smooth communication between document indexing, retrieval, and answer generation. The backend tools can be extended or replaced easily as new functionalities are added. -
PDF Upload and Parsing:
Upload a PDF file which is then parsed usingPDFPlumberLoaderto extract text content. -
Text Chunking:
The extracted text is split into smaller chunks usingRecursiveCharacterTextSplitterto facilitate efficient indexing and retrieval. -
Document Indexing:
Chunks are indexed in an in-memory vector store with embeddings generated viaOpenAIEmbeddings. -
Similarity Search (Consine Similarity):
When a user submits a query, the chatbot performs a similarity search to retrieve the most relevant document chunks based on the query. -
Prompt-Based Answer Generation:
A custom prompt template combines the user question with retrieved context, and a GPT-4 powered LLM (usingChatOpenAI) generates the final answer. -
Interactive Interface:
The application uses Streamlit to provide an interactive, chat-like interface where user questions and bot responses are displayed.
Result
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.