Cabeza

Your AI has amnesia. Cabeza fixes that. Cabeza gives your favorite LLM, AI agnostic, AWARENESS of facts, decisions, and memories established earlier in conversation that have fallen out of the context window, so you never have to tell your AI "we already discussed this" again. Every conversation you have with ChatGPT, Claude, or Gemini is automatically captured as clean markdown and stored in your own *private* Google Drive or Dropbox. Not our servers — yours. No setup. No prompts. No system to manage. It just works. HOW IT WORKS • Install the Cabeza extension and connect your Google Drive or Dropbox • Use ChatGPT, Claude, or Gemini like you normally do • Cabeza silently captures every conversation in the background • Search any past conversation instantly with Alt+Shift+M • Your AI automatically gets context from previous conversations WHAT MAKES CABEZA DIFFERENT • Cross-platform memory — one memory across ChatGPT, Claude, and Gemini • Your storage, your data — everything stays in your Drive or Dropbox • Automatic capture — no buttons to click, no effort required • Smart context injection — your AI remembers who you are and what you've discussed • Long conversation support — re-injects earlier context when conversations drift. No more re-explaining to AI what you were talking about. It just remembers. • On-device ML — semantic search powered by local AI, no API keys needed PRIVACY FIRST Your conversations are stored in YOUR Google Drive or Dropbox. They never touch Cabeza servers — because there are no Cabeza servers. The extension runs entirely in your browser. Your data flows directly between Chrome and your storage provider. PRICING Free to install. Try everything first. Lock in a $0.99 one-time unlock before pricing switches to $5+/month. Early adopters are grandfathered permanently. PERMISSIONS EXPLAINED • Storage access: saves settings and caches ML models locally • Identity: connects to your Google Drive without extra login steps • Active tab & scripting: reads conversations on supported AI platforms • Context menus: right-click to search or save memory • Offscreen: runs ML models off the main thread so your browser stays fast Built for desktop — where real work happens. More info: https://cabeza.cc