A powerful CLI tool that uses Google's Gemini AI to generate, iterate upon, and export professional PDF documents (Resumes, Proposals, Reports, etc.) directly from your terminal.
- Interactive AI Chat: Brainstorm and edit your document in real-time.
- Rich Console Rendering: View beautifully formatted Markdown in your terminal before exporting.
- Professional PDF Styling: Converts Markdown to styled PDFs using WeasyPrint.
- Persistent History: The AI remembers your previous edits, allowing for iterative document building.
Before installing, you must have the following graphics libraries installed on your system (required for PDF rendering):
brew install pango libffi
Linux (Ubuntu/Debian)
sudo apt-get install libpango-1.0-0 libpangoft2-1.0-0 libffi-dev
🚀 Installation
Set your Gemini API Key:
Obtain a key from Google AI Studio and add it to your shell profile:
export GEMINI_API_KEY="your_api_key_here"
Install the package:
Navigate to the project root and run:
pip install .
⌨️ Usage
Simply type the following command to launch the architect:
Bash
ai-docCommands within the app:
Just type anything: Talk to the AI to generate or edit content.
/pdf [filename.pdf] : Export the current text to a styled PDF file.
/exit : Close the application.
📂 Project Structure
ai_doc/ : Core Python package containing the logic.
pyproject.toml : Build system requirements and CLI entry points.
main.py : Interactive loop and API integration.