Command Palette

Search for a command to run...

UnylyUnyly
Browse all

File Assistant

FreeNot checked

MCP server that allows AI agents to read, search, summarize, and answer questions about local PDF, DOCX, XLSX, and TXT files through five tools.

GitHubEmbed

About

MCP server that allows AI agents to read, search, summarize, and answer questions about local PDF, DOCX, XLSX, and TXT files through five tools.

README

A document intelligence system that enables AI agents and users to read, search, summarize, and answer questions over local PDF, DOCX, XLSX, and TXT documents.

The project exposes the same core engine through:

  • MCP Server for Claude Desktop and other MCP-compatible clients
  • Flask Web Application for browser-based interaction

Features

  • Read PDF, DOCX, XLSX and TXT documents
  • Search keywords across multiple documents
  • AI-powered document summarization
  • Retrieval-Augmented Document Q&A
  • Upload and manage files
  • Delete individual or all documents
  • Secure file access with path validation
  • Claude Desktop MCP integration

Tech Stack

Category Technologies
Backend Python, Flask
MCP Model Context Protocol (MCP)
AI Groq Llama / Ollama
Document Parsing PyMuPDF, python-docx, pandas, openpyxl
Frontend HTML, CSS, JavaScript
Testing Pytest

Architecture

                 Claude Desktop
                       │
                       ▼
                 MCP Server
                       │
        ┌──────────────┴──────────────┐
        │                             │
   Shared Core Engine            Flask Web App
        │
 ├── File Readers
 ├── Search Engine
 ├── Document Processor
 └── LLM Client

📸 Screenshots

🖥️ User Interface

Modern web interface for uploading, browsing, and managing supported documents.

image

🔗 MCP Server Connected

Claude Desktop successfully connected to the local MCP File Assistant server.

image

📄 Read Document

Read PDF, DOCX, XLSX, and TXT documents directly through Claude Desktop using MCP.

image image

📝 AI Document Summarization

Generate concise AI-powered summaries of lengthy documents using Groq Llama.

image

🔍 Keyword Search

Search keywords across every indexed document with ranked search results.

image

❓ Document Question Answering

Ask natural language questions grounded in document contents using Retrieval-Augmented Generation (RAG).

image

🗑️ File Management

Delete individual files or remove all uploaded documents directly from the application.

image

Claude Desktop MCP Integration

Claude Desktop successfully invokes custom MCP tools including Read, Search, Summarize, and Ask Document.

image

📁 Project Structure

mcp-file-assistant/
│
├── app.py
├── server.py
├── config.py
├── file_readers.py
├── search_engine.py
├── document_processor.py
├── llm_client.py
├── templates/
├── static/
├── tests/
├── sample_documents/
├── requirements.txt
└── README.md

🚀 Installation

git clone <repository-url>

cd mcp-file-assistant

pip install -r requirements.txt

python app.py

Open:

http://127.0.0.1:5000

To start the MCP server:

python server.py

Configure Claude Desktop with the provided MCP configuration and restart the application.


🔧 MCP Tools

  • Read Document
  • Search Documents
  • Summarize Document
  • Ask Document
  • List Files
  • Delete File
  • Delete All Files

from github.com/Kaviyasree27/MCP-File-Assistant-with-Claude-Desktop-integration

Installing File Assistant

This server has no published package — it is built from source. Open the repository and follow its README.

▸ github.com/Kaviyasree27/MCP-File-Assistant-with-Claude-Desktop-integration

FAQ

Is File Assistant MCP free?

Yes, File Assistant MCP is free — one-click install via Unyly at no cost.

Does File Assistant need an API key?

No, File Assistant runs without API keys or environment variables.

Is File Assistant hosted or self-hosted?

Self-hosted: the server runs locally on your machine via the install command above.

How do I install File Assistant in Claude Desktop, Claude Code or Cursor?

Open File Assistant on unyly.org, pick your client tab (Claude Desktop, Claude Code, Cursor) and press Install — the config is generated automatically, no JSON editing.

Related MCPs

Compare File Assistant with

Not sure what to pick?

Find your stack in 60 seconds

Author?

Embed badge for your README

Browse similar

All ai MCPs