Command Palette

Search for a command to run...

UnylyUnyly
Browse all

Agent Materials

FreeNot checked

πŸ“š Materials for a data analysis agent using YouTube MCP

GitHubEmbed

About

πŸ“š Materials for a data analysis agent using YouTube MCP

README

이 μ €μž₯μ†ŒλŠ” AI Agent 개발, 데이터 μˆ˜μ§‘ 및 뢄석, 그리고 챗봇 μ œμž‘ 및 배포에 이λ₯΄λŠ” 일련의 과정을 μ‹€μŠ΅ν•  수 μžˆλ„λ‘ κ΅¬μ„±λ˜μ–΄ μžˆμŠ΅λ‹ˆλ‹€.

μ΅œμ’… 결과물은 μ•„λž˜ λ ˆν¬μ§€ν† λ¦¬μ—μ„œ ν™•μΈν•˜μ‹€ 수 μžˆμŠ΅λ‹ˆλ‹€.

MCP Server

MCP Client

ν™˜κ²½ μ„€μ • 및 ν•„μˆ˜ 파일

파일/폴더 μ„€λͺ… μ„€μ • 방법
.env.example* ν™˜κ²½ λ³€μˆ˜ μ„€μ • 파일의 μ˜ˆμ‹œμž…λ‹ˆλ‹€. 이 νŒŒμΌμ„ λ³΅μ‚¬ν•˜μ—¬ .env* νŒŒμΌμ„ μƒμ„±ν•˜κ³ , YouTube 및 Gemini API Keyλ₯Ό λ°œκΈ‰λ°›μ•„ λΆ™μ—¬λ„£μœΌμ„Έμš”.
requirements.txt* ν”„λ‘œμ νŠΈμ— ν•„μš”ν•œ 라이브러리 λͺ©λ‘μž…λ‹ˆλ‹€. λ‹€μŒ μ½”λ“œλ₯Ό μ‚¬μš©ν•˜μ—¬ 가상 ν™˜κ²½μ„ μ„€μ •ν•˜κ³  νŒ¨ν‚€μ§€λ“€μ„ λ‹€μš΄λ‘œλ“œν•˜μ„Έμš”:
python -m venv venv
source venv/bin/activate (Linux/macOS) λ˜λŠ” .\venv\Scripts\activate (Windows)
pip install -r requirements.txt

πŸ“š μ‹€μŠ΅ 자료 ꡬ성 (4단계 μ›Œν¬ν”Œλ‘œμš°)

1. streamlit 폴더: AI Agent κ°œλ… 및 개발 ν™˜κ²½ μ„ΈνŒ…

AI Agent의 기초 지식 μŠ΅λ“κ³Ό 개발 ν™˜κ²½ 섀정을 λͺ©ν‘œλ‘œ ν•©λ‹ˆλ‹€.

  • 1-1 AI Agent 기초: AI Agent의 κΈ°λ³Έ κ°œλ… ν•™μŠ΅

  • 1-2 개발 ν™˜κ²½ μ€€λΉ„: ν•„μš”ν•œ 도ꡬ 및 ν™˜κ²½ μ„€μ •

  • 1-3 Streamlit 기초: μ›Ή μ• ν”Œλ¦¬μΌ€μ΄μ…˜ ν”„λ ˆμž„μ›Œν¬ Streamlit κΈ°λ³Έ μ‚¬μš©λ²•

  • 1-4 Gemini API μ—°λ™ν•˜κΈ°: Gemini APIλ₯Ό ν™œμš©ν•œ 연동 방법 (GPT API μ†Œκ°œ 포함)

  • 1-5 μ‹€μŠ΅: κΈ°λ³Έ 챗봇 개발 (Gemini API μ—°λ™ν•˜μ—¬ Streamlit으둜 λ„μ›Œλ³΄κΈ°)

2. mcp 폴더: 데이터 μˆ˜μ§‘

YouTube 데이터 μˆ˜μ§‘μ— ν•„μš”ν•œ 기초 지식 및 MCP(Multi-Channel Processing) ν™œμš©λ²•μ„ λ‹€λ£Ήλ‹ˆλ‹€.

  • 2-1 데이터 μˆ˜μ§‘ 기초: 데이터 μˆ˜μ§‘μ˜ κΈ°λ³Έ κ°œλ…

  • 2-2 MCPλ₯Ό ν™œμš©ν•œ 유튜브 데이터 μˆ˜μ§‘: MCP 도ꡬλ₯Ό μ‚¬μš©ν•œ 유튜브 데이터 μˆ˜μ§‘ 방법

  • 2-3 μ‹€μŠ΅: API와 MCPλ₯Ό ν™œμš©ν•œ 유튜브 데이터 μˆ˜μ§‘

  • 2-4 데이터 슀트리밍 & μŠ€μΌ€μ€„λ§: μ‹€μ‹œκ°„ 데이터 처리 및 μž‘μ—… μžλ™ν™” κ°œλ…

  • 2-5 μ‹€μŠ΅: 데이터 μŠ€μΌ€μ€„λ§μ„ ν™œμš©ν•œ 유튜브 데이터 μˆ˜μ§‘

3. data 폴더: 데이터 뢄석

μˆ˜μ§‘λœ 데이터λ₯Ό νƒμƒ‰ν•˜κ³  λΆ„μ„ν•˜λŠ” 과정을 ν”„λ‘¬ν”„νŠΈ μ—”μ§€λ‹ˆμ–΄λ§κ³Ό ν•¨κ»˜ μ‹€μŠ΅ν•©λ‹ˆλ‹€.

  • 3-1 데이터 뢄석 기초: 데이터 λΆ„μ„μ˜ κΈ°λ³Έ 원리 및 방법

  • 3-2 μ‹€μŠ΅: **탐색적 데이터 뢄석(EDA)**와 ν…μŠ€νŠΈ μ „μ²˜λ¦¬

  • 3-3 데이터 뢄석기: 데이터 뢄석을 μœ„ν•œ 도ꡬ 및 라이브러리 μ†Œκ°œ

  • 3-4 μ‹€μŠ΅: ν”„λ‘¬ν”„νŠΈ μ—”μ§€λ‹ˆμ–΄λ§ 기초 및 적용

  • 3-5 μ‹€μŠ΅: λ‰΄μŠ€ 데이터 μžλ™ 뢄석 & μš”μ•½ 슀크립트 μ œμž‘

4. deploy 폴더: 챗봇 μ œμž‘ 및 배포

κ°œλ°œν•œ 챗봇을 μ™„μ„±ν•˜κ³  μ‹€μ œ ν™˜κ²½μ— λ°°ν¬ν•˜λŠ” 과정을 μ‹€μŠ΅ν•©λ‹ˆλ‹€.

  • 4-1 챗봇 ν™”λ©΄ 섀계: μ‚¬μš©μž μΉœν™”μ μΈ 챗봇 μΈν„°νŽ˜μ΄μŠ€ 섀계

  • 4-2 배포의 κ°œλ…: 개발된 μ• ν”Œλ¦¬μΌ€μ΄μ…˜μ„ μ„œλΉ„μŠ€ν•  수 있게 λ§Œλ“œλŠ” κ³Όμ • ν•™μŠ΅

  • 4-3 μ‹€μŠ΅: λ‰΄μŠ€ νŠΈλ Œλ“œ 챗봇 μ™„μ„± 및 배포


AI Agent and Data Analysis Workshop Materials

This repository is designed for hands-on practice covering the entire workflow of AI Agent Development, Data Collection & Analysis, and Chatbot Creation & Deployment.

You can find the final deliverables in the repositories below:

MCP Server

MCP Client

Environment Setup & Essential Files

File/Folder Description Setup Method
.env.example An example file for environment variable configuration. Copy this file to create a .env file, then issue and paste your YouTube and Gemini API Keys.
requirements.txt A list of libraries required for the project. Use the following code to set up a virtual environment and download the packages:
python -m venv venv
source venv/bin/activate (Linux/macOS) or .\venv\Scripts\activate (Windows)
pip install -r requirements.txt

Workshop Structure (4-Stage Workflow)

1. streamlit folder: AI Agent Concepts & Development Environment Setup

Aims to acquire basic knowledge of AI Agents and set up the development environment.

  • 1-1 AI Agent Basics: Learning the fundamental concepts of AI Agents
  • 1-2 Environment Preparation: Setting up necessary tools and environments
  • 1-3 Streamlit Basics: Basic usage of the web application framework, Streamlit
  • 1-4 Gemini API Integration: How to integrate using the Gemini API (includes GPT API introduction)
  • 1-5 Lab: Developing a Basic Chatbot (Integrating Gemini API and launching it with Streamlit)

2. mcp folder: Data Collection

Covers the basic knowledge required for YouTube data collection and how to utilize MCP (Multi-Channel Processing).

  • 2-1 Data Collection Basics: Fundamental concepts of data collection
  • 2-2 YouTube Data Collection with MCP: How to collect YouTube data using MCP tools
  • 2-3 Lab: YouTube data collection using API and MCP
  • 2-4 Data Streaming & Scheduling: Concepts of real-time data processing and task automation
  • 2-5 Lab: YouTube data collection using Data Scheduling

3. data folder: Data Analysis

Practice exploring and analyzing collected data alongside Prompt Engineering.

  • 3-1 Data Analysis Basics: Basic principles and methods of data analysis
  • 3-2 Lab: Exploratory Data Analysis (EDA) and Text Preprocessing
  • 3-3 Data Analyzer: Introduction to tools and libraries for data analysis
  • 3-4 Lab: Prompt Engineering Basics & Application
  • 3-5 Lab: Creating an Automated News Data Analysis & Summary Script

4. deploy folder: Chatbot Creation & Deployment

Practice finalizing the developed chatbot and deploying it to a live environment.

  • 4-1 Chatbot UI Design: Designing a user-friendly chatbot interface
  • 4-2 Deployment Concepts: Learning the process of making a developed application serviceable
  • 4-3 Lab: News Trend Chatbot Completion & Deployment

from github.com/2shin0/mcp-agent-materials

Installing Agent Materials

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

β–Έ github.com/2shin0/mcp-agent-materials

FAQ

Is Agent Materials MCP free?

Yes, Agent Materials MCP is free β€” one-click install via Unyly at no cost.

Does Agent Materials need an API key?

No, Agent Materials runs without API keys or environment variables.

Is Agent Materials hosted or self-hosted?

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

How do I install Agent Materials in Claude Desktop, Claude Code or Cursor?

Open Agent Materials on unyly.org, pick your client tab (Claude Desktop, Claude Code, Cursor) and press Install β€” the config is generated automatically, no JSON editing.

Related MCPs

Compare Agent Materials with

Not sure what to pick?

Find your stack in 60 seconds

Author?

Embed badge for your README

Browse similar

All media MCPs