Activity log

On web and desktop, consecutive tool calls appear as an expandable summary. Open it to inspect commands, tool inputs, status, and exit codes. Expand a tool call to see its output. A running command's output updates live, and long output shows only its most recent part. Use Open diff on a file change to review its contents.

T3 Orchestrator summaries describe the actions taken, such as Ran 2 commands and sent messages to 3 threads. Repeated messages to the same destinations show both counts: Sent 5 messages to 2 threads. Thread creation counts the threads returned by the tool, including batches.

Groups with many kinds of activity show up to two specific action categories and a count of the remaining actions. Commands, file changes, and orchestration changes take priority over reads and status checks. Expand the group for the full list.

Failed calls do not count as successful messages or creations. Waiting on a thread does not mean it finished, and an interrupt or cancellation request does not mean the thread stopped. When tool details are unavailable, summaries use a broader description instead of guessing how many threads were affected.

Hide tool calls

Tool calls in chat in Settings → Chat (web and desktop) or Settings → Appearance (mobile) hides commands, file edits, searches, and MCP tool calls from the conversation. Hide successful shows a call while it runs and keeps it only if it fails. Hide all hides every call and shows Thinking while the agent works. Zen also hides the agent's in-between messages, so each turn shows your message and the agent's last reply, which appears when the turn ends. If you steer while the agent is answering, the answer it finished before reading your steer stays too. On web and desktop, Find shows the in-between messages while it is open. Approvals, questions, and errors always show. The setting applies to this browser or device only.

Screen reader announcements

On web and desktop, a screen reader hears the open thread's turn as it runs. Screen reader announcements in Settings → Chat sets how much. Status only says when a turn starts, needs your approval or input, and ends. Status and messages also reads each message the agent writes and then keeps working after, once that message is complete. A very long message is cut short. The reply that closes the turn is not read, so you can read it yourself when you hear Done. Everything adds the step the agent is on, such as the command it is running. Steps are spaced out, so a quick step can be skipped for a newer one. The agent's reasoning text is never read. The setting follows the agent's real activity, whatever Tool calls in chat hides, and applies to this browser or device only.