Donkey

Persistent memory layer across AI tools

Donkey gives your AI conversations a persistent memory — so you never have to re-explain your project from scratch. Most AI tools forget everything the moment a chat ends. Donkey fixes that. It lets you save the full context of any conversation — decisions made, mistakes to avoid, learnings, constraints, open questions — and recall exactly the right pieces in your next session with a simple command. HOW IT WORKS Type @donkey save at the end of any conversation to store the entire thread as a structured project memory. Donkey extracts decisions, learnings, mistakes, and context automatically. To continue work in a new session, type @donkey use <project-name> and Donkey injects the saved memory directly into your chat box — ready to send. Need just a slice? Use filters like @donkey learnings from <project> or @donkey mistakes from <project> to pull in only the relevant parts. COMMANDS • @donkey save — Saves the active chat thread as a named project memory • @donkey use <project> — Injects the full saved context for a project into your chat • @donkey <filter> from <project> — Pulls specific segments (learnings, mistakes, decisions, constraints, context, open questions) SUPPORTED PLATFORMS Works on Claude, ChatGPT, and Gemini — switch between AI tools without losing context. PRIVACY All memory is stored locally on your device using browser storage. Nothing is sent to external servers. Your conversations stay yours. PLANS Free plan available. Lifetime access for $29 — one-time payment, no subscription.