Run your agents in parallel, direct them by voice, and pick the work up from anywhere. Each habit makes the others stronger, together they multiply how much you get done. Under the hood it's a native app, speaking real tmux control mode. One session follows you across Mac, iPhone and iPad; your agents run on macOS, Linux or WSL. No account required. Voice works out of the box, or bring your own API key. Free & open source.
Wispr Flow: Dictation That Works EverywhereStop typing. Start speaking. 4x faster.
Promoted
Maker
📌
Hi Product Hunt! I run several Claude Code sessions side by side in tmux. Originally I just wanted a tmux client for my iPhone and iPad — split panes plus voice input, so I could keep working while looking after my kid. Then I noticed I was getting more done on the iPad than at my 27-inch monitor, so I built the Mac version too. Three months later it's my daily driver.
The way of working is three things that reinforce each other: run agents in parallel, direct them by voice, and pick the work up from anywhere. Every agent gets a pane on one screen, so a glance tells you who needs attention. You answer by talking — the recognition reads the screen, so file names and identifiers come out right. And because everything lives in real tmux over SSH, the same session follows you from the Mac to your phone.
Under the hood: native Swift, libghostty rendering (Ghostty's engine), real tmux control mode. Nothing installed on your remote machine. No account, bring your own API key if you like. Free and Apache-2.0.
Would love your feedback — especially from anyone juggling long-running agents.