Personal AI Memory is an open-source Chrome extension that captures your AI conversations and builds a searchable, private memory graph. It supports major platforms like OpenAI's ChatGPT, Google's Gemini, Anthropic's Claude, and Perplexity, ensuring your prompt engineering history is safely stored and easily accessible.
The extension automatically captures conversations from ChatGPT and Gemini Web UI by parsing the DOM to capture and index your thoughts without changing your workflow. It provides instant recall capabilities, allowing you to easily import and export conversations and inject historical context into your next prompt with a single click. The tool features 100% local privacy with no servers or API keys required, keeping all data in IndexedDB.
Personal AI Memory uses a local-first architecture powered by WebAssembly on the edge. It works silently in your browser without interrupting your workflow, capturing conversations locally and building a searchable memory graph.
The extension helps users master prompt engineering across multiple AI models by preventing fragmented context where conversations are trapped in separate silos. It allows you to instantly recall past context and inject it directly into new prompts on any supported platform, making it easier to find conversations from weeks ago without digging through multiple platform histories.
The target users include anyone using multiple AI platforms like ChatGPT for coding, Claude for writing, and Perplexity for research. The extension integrates directly with browser interfaces of supported AI platforms and is built using Manifest V3 and WebAssembly technology.
admin
Personal AI Memory is designed for users who work with multiple AI platforms like ChatGPT for coding, Claude for writing, and Perplexity for research. It targets individuals who need to manage their prompt engineering history across different services and want to prevent fragmented context where conversations are trapped in separate silos. The extension appeals to privacy-conscious users who prefer local-first architecture and want to avoid sharing their data with third-party servers.