AI_interaction is a modern MCP server that provides an interactive UI popup for content input and conversation control, with built-in support for attaching files, folders, and multiple images. The tool outputs data using a structured, tag-based format that integrates with your system prompt rules, making it easy to influence AI behavior through tags and contextual data. You can attach workspace files and folders, drag and drop images, and switch between languages (English and Vietnamese) to suit your environment. Use the UI to compose rich prompts, manage attachments, and receive responses that respect the predefined tag schema for downstream agents or workflows.

Prerequisites:

• Python 3.8 or newer
• Git
• Access to install Python dependencies from requirements.txt

Installation steps:

1. Clone the repository: git clone https://github.com/your-username/AI-interaction.git cd AI-interaction

2. Create and activate a Python environment (optional but recommended): python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate

3. Install dependencies: pip install -r requirements.txt

   If you prefer using uv for performance (optional):

   pip install uv uv pip install -r requirements.txt

4. Run the MCP server locally to test: python mcp_server.py

5. Optional: Configure Claude Desktop or Cursor with the MCP server settings as shown in the README to enable external tooling integration.

Tips and common issues:

• Ensure the path to mcp_server.py is correct when configuring external clients (Claude Desktop, Cursor, etc.).
• If you enable image attachments, ensure your user_images directory has appropriate write permissions.
• The tool uses a tag-based output format; familiarize yourself with the tag rules to leverage advanced AI agent processing.
• For performance, consider running the server with uv (as shown in installation) if you expect high concurrency.
• If you update dependencies, re-run pip install -r requirements.txt to keep the environment in sync.