Ofrenda 🌼 — The Offline Memory Keeper | #DEVChallenges #HF26

#devchallenge #hf26challenge #opensource #python

Hacktoberfest Open-Source AI Challenge: Ghost in the Machine Submission 👻🌼

🌼 Ofrenda — The Offline Memory Keeper

A local-first AI memory keeper for preserving family stories, cherished recipes, personal memories, and wisdom.

🔗 GitHub Repository: https://github.com/vanshikamishra1909-cmd/OFRENDA-THE-MEMORY-KEEPER

💀 The Idea

Some memories are too precious to lose.

Inspired by the tradition of ofrendas—altars honoring and remembering loved ones—Ofrenda brings that spirit into a small, local-first AI application. It helps people preserve meaningful stories, family recipes, and personal wisdom in a searchable digital memory vault.

The goal is simple: make it easier to keep memories close while reducing reliance on hosted AI services.

✨ What It Can Do

📝 Memory Vault: Add, browse, and delete saved memories.
🔎 Memory Search: Retrieve relevant records using lightweight keyword matching.
💬 Ask the Archive: Ask questions grounded in retrieved memories using a locally running Ollama model.
🕯️ Calaveritas Studio: Generate remembrance poems inspired by selected memories.
📦 JSON Export: Export memory records for safekeeping.
💻 Local-First Storage: Store records in a local SQLite database and use Ollama for AI generation.

🛠️ Tech Stack

Python
Streamlit
SQLite
Ollama
Qwen 2.5 3B (default local model)

🚀 Run It Locally

Requirements: Python 3.10+, Ollama, and a locally downloaded model.

Clone the repository:

git clone https://github.com/vanshikamishra1909-cmd/OFRENDA-THE-MEMORY-KEEPER.git
cd OFRENDA-THE-MEMORY-KEEPER

Create and activate a virtual environment on Windows:

py -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
ollama pull qwen2.5:3b

Make sure Ollama is running, then launch the application:

streamlit run app.py

Open the local URL shown in your terminal, usually http://localhost:8501.

🔐 Privacy Matters

Ofrenda is designed as a local-first prototype. Its default configuration stores memories in SQLite and sends prompts to the local Ollama API at 127.0.0.1, rather than intentionally calling a hosted AI API.

However, this is an early prototype—not a security-audited product. The database is not encrypted, and there is no login or multi-user access control. Please review the privacy limitations in the repository before importing sensitive information.

🧪 Project Status

Ofrenda is an early MVP. It currently supports text memories, keyword-based retrieval, local-model question answering, remembrance poem generation, and JSON export. Semantic search, image gallery support, and audio transcription are not implemented.