Built an AI Voice-Operated Notification Summarizer for #Hacktoberfest2026 using Flutter, Open-Source Gemma & Gemini MCP!

Drowning in daily notifications? For Hacktoberfest 2026, I developed an open-source Flutter app that does more than just summarize your notifications—it enables you to manage them hands-free with a local AI voice assistant. Unlike cloud-based solutions, this project keeps your private phone notifications secure by using a hybrid AI architecture that blends on-device privacy with real-time internet intelligence.

The app's core is an open-source Gemma model, which acts as the primary brain. It reads your incoming notifications, categorizes them by priority (Work, Social, Updates), and creates a concise daily summary. The app allows you to interact with your notification log through voice commands. For example, you can say "Show urgent alerts," "Mark work messages as read," or "Delete all promotional updates," and Gemma translates these voice intents into real-time Create, Read, Update, and Delete (CRUD) operations on a local database.

In addition to Gemma, the app integrates the Gemini Free API along with a Model Context Protocol (MCP) for internet fallback. This means you can ask the voice assistant any question. If the query requires live world knowledge or something the local Gemma model doesn't know, it automatically routes the request through MCP using a free-tier Gemini API key to search the live web and provide the answer verbally.

The tech stack includes Flutter for the cross-platform UI, with built-in speech-to-text and text-to-speech capabilities. The local orchestrator runs on Gemma, an open-source LLM, handling intent parsing, summarization, and CRUD tool calling. For live internet access, a free-tier Gemini API, connected via MCP, is used as a fallback.

The project is officially open for Hacktoberfest contributions. Whether you want to add new MCP tools, refine the Gemma prompt templates, or enhance the Flutter UI, check out the repository and grab an issue.