Paint Drawing Agent

Name: Paint Drawing Agent
Author: shivman

@shivman

Paint Drawing Agent について

This project demonstrates an Agentic AI system that uses a Gemini-powered MCP server to interpret natural language instructions and automate Microsoft Paint using GUI simulation tools on Windows.

#paint-ai-agent #automation #ai

基本情報

カテゴリ

AI とエージェント

ランタイム

python

トランスポート

stdio

公開者

shivman

設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "paint-ai-agent": {
      "command": "python",
      "args": [
        "talk2mcp.py"
      ]
    }
  }
}

ツール

ツールは検出されませんでした

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Paint Drawing Agent?

A Python-based automation tool that uses Google's Gemini AI to control Microsoft Paint through natural language commands, enabling users to draw shapes, write text, and manipulate colors in MS Paint using simple English instructions.

How to use Paint Drawing Agent?

Clone the repository, install dependencies with pip install -r requirements.txt, create a .env file with your Google API key, then run python talk2mcp.py and enter natural language commands like "Draw a red circle in the center". Type 'quit' to exit.

Key features of Paint Drawing Agent

Natural language control of Microsoft Paint
Automated drawing of circles, rectangles, and lines
Text insertion with position control
Color selection and management
Window management and canvas positioning
Detailed logging and error handling
Position calibration system for accuracy

Use cases of Paint Drawing Agent

Automating repetitive drawing tasks in MS Paint
Creating diagrams or annotations without manual mouse interaction
Generating custom shapes and text for quick visual mockups

FAQ from Paint Drawing Agent

What prerequisites are required?

Windows 10 or later, Python 3.8+, Microsoft Paint (mspaint.exe), and a Google Cloud API key for Gemini AI.

How does calibration work?

The system uses a calibration system to accurately locate Paint interface elements. Run python enhanced_calibrate.py and follow on-screen instructions to recalibrate tool positions.

Where are logs stored?

Application logs are in /logs/paint_agent_[timestamp].log, LLM interaction logs in /LLM_LOGS/session_log.json, and calibration data in /calibration_profiles/.

What should I do if Paint is not found?

Ensure MS Paint is installed and try running Paint manually first. If drawing positions are off, run calibration again and check that screen resolution matches the calibration profile.