How my AI assistant thinks — an interactive map

A walkable diagram of how my memory-equipped AI assistant actually works: its senses, its memory, the mind that composes each reply, its hands that act, how it reaches out on its own, how it knows itself, and how each day settles back into memory. Every part reads from and writes to the Memory at the center. Watch the particles — they show which way data actually moves. Pick a loop to trace one path, or click a node to open it.

Trace a loop:
The userSensesMemoryThe MindVoiceReaching outSleep & LearningCognitive CoreSelf-knowledgeActing — its hands
The userby text or voiceSensesthe world gets inMemoryits mind, on diskThe Mindcomposes every replyVoicetext becomes speechReaching outit starts conversationsSleep & Learninga day becomes memoryCognitive Coreanticipates & prioritizesSelf-knowledgeit understands itselfActing — its handsit changes things, not just talks
The whole map

Everything, connected

Every part reads from and writes to the Memory at the center. Senses bring the world in; the Mind turns it into a reply; our assistant's hands act on the world; it reaches out on its own; it holds a model of itself; and every conversation — every few minutes — settles back into memory as durable facts and its own views. Pick a loop above to trace one path, or click any node to open it.

Tip: watch the particles — they show which way data actually moves.
    Interactive system map · how our AI assistant actually works, as an animated data-flow diagram. Plain-language; built as a single self-contained HTML file (SVG + vanilla JS, no dependencies).