Chat
A box is a directory of your data, kept under version control with a full history of changes (using git); a card is a markdown file with a structured header; the agent is the coding agent (Claude Code or Codex) that operates the box. Chat is the everyday way you talk to it.
What it does for you
- Streams the agent's reply as it is generated, instead of making you wait for the whole answer.
- Lets you resume a past conversation, reopen a chat from its card, or pick up the most recently active one.
- Scopes a chat to wherever you started it: a card you were reading, a directory ("place") in the box, or a page you just clipped from the web.
- Lets you quote a passage out of any card into what you're typing.
- Recovers a message that failed to send instead of losing it, and keeps an unfinished draft across reloads.
- Can set a reminder mid-conversation that wakes the same chat back up later, optionally with a sound or a spoken announcement.
What it needs
Nothing beyond the box itself and a working agent login (Claude Code or Codex). See ../install/index.md.
How it works, briefly
Every conversation is a durable chat card, so it survives reloads and is browsable like anything else in the box. The agent acts inside a turn and asks you a question when it is unsure rather than guessing; its mid-turn work (files touched, background notes) is visible in the transcript. Chat runs on demand, when you send a message, not on a schedule.
Limits
A chat that cannot be resumed (for example, an old session format) says so rather than silently starting a new, disconnected one. The documentation does not describe a message-editing feature; a sent message stands as sent.
Go deeper
../reference/cards/chat.md, ../reference/cards/chat-thread.md, ../reference/chat-voice.md, ../reference/narration-mode.md, voice.md