AI-Powered Todo App

A modern, Notion-inspired AI Todo app built with Streamlit, MongoDB, and Gemini (Google Generative AI).

Features

• Beautiful UI: Notion-like dark theme, compact cards, and modern design.
• AI Assistant: Manage todos using natural language (powered by Gemini via LangChain).
• CRUD Operations: Add, list, complete, modify, and delete todos.
• Summarization: Get AI-generated summaries of your todo list.
• Voice Input: (Optional) Use browser-based voice-to-text in Chrome.
• MongoDB Backend: Todos are stored in a MongoDB Atlas database.
• Per-user/device support: (Planned) Each user/device can have their own todo list.

Quickstart

1. Clone the repository

git clone <your-repo-url>
cd python-todo

2. Install dependencies

pip install -r requirements.txt

3. Set up environment variables

Create a .env file (for local dev) or use Streamlit Cloud secrets for deployment:

MONGODB_URI="mongodb+srv://<username>:<password>@<cluster>.mongodb.net/<dbname>"
GEMINI_API_KEY="<your-gemini-api-key>"
GOOGLE_API_KEY="<your-google-api-key>"

4. Run the app locally

streamlit run app/main.py

5. Deploy to Streamlit Cloud

• Push your code to GitHub.
• Go to Streamlit Cloud and create a new app.
• Set your main file to app/main.py.
• Add your secrets (MongoDB URI, API keys) in the Secrets manager.

MongoDB Atlas Setup

• Create a free cluster at MongoDB Atlas.
• Add a database user with read/write access.
• In Network Access, add 0.0.0.0/0 to allow access from anywhere (for testing).
• Copy your connection string and use it as MONGODB_URI.

AI & Voice

• Gemini (Google Generative AI) is used for natural language todo management and summarization.
• Voice input uses the browser's SpeechRecognition API (works best in Chrome).

File Structure

python-todo/
├── app/
│   ├── main.py           # Streamlit UI
│   ├── crud.py           # CRUD operations
│   ├── database.py       # MongoDB connection
│   ├── models.py         # Pydantic models
│   └── ai/
│       ├── nlp.py        # AI logic
│       ├── llm.py        # Gemini LLM wrapper
│       ├── prompts.py    # Prompt templates
│       ├── tools.py      # Tool-calling wrappers
│       └── chains.py     # (Optional) LangChain chains
├── requirements.txt
├── .env (local only)
└── README.md

Troubleshooting

• ModuleNotFoundError: Make sure all dependencies are in requirements.txt.
• MongoDB connection errors: Check your URI, credentials, and Atlas network access.
• Secrets not working: Use Streamlit Cloud's Secrets manager, not .env.
• Voice input not working: Use Chrome and allow microphone access.

Credits

• Built with Streamlit, MongoDB Atlas, LangChain, and Gemini.

---

MIT License