Say So: An Open-Source AI Assistant for the Replies You Keep Putting Off
This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady What I Built Have you ever opened a WhatsApp message, read it, and then left it there—not because you don't care about the person, but because you just don't have the energy to reply? Sometimes, you're overstimulated. You've already spent the day processing information, responding to people, and thinking…
This story is about an open-source AI assistant called "Say So" designed for WhatsApp Web. The creator built it to address two common issues people face when replying to messages: personalized replies for direct messages and summaries for unread group chats. Say So uses conversation history, the current message, and context to build contact-specific communication profiles.
These profiles help generate replies that are relevant to the conversation and feel natural to the sender. Instead of blindly copying old messages, Say So presents three reply options for the user to choose from or edit. Additionally, the app can summarize unread group chats into concise overviews, key points, and action items. The open-source nature of Say So allows developers to inspect, test, and improve the application.
It is built using a TypeScript-based Chrome extension, a backend that connects WhatsApp Web to AI-powered functionality, and various technologies such as MongoDB Atlas, Supabase Auth, Render, and Sentry.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.