Tags
VOICEVoice-driven production. CAPTUREScreen, GIF, scrolling capture or clipboard signal. ANNOTATIONMark, number, box or arrow tied to a speech moment. WORKFLOWRepeatable working flow.
1 — Start recording and speak naturally. Do not try to make the first sentence perfect. The live waveform shows that audio is being received, and the small progress indicator shows that speech is becoming text in the background.
1 2 3 4 5
Captured items — the stack grows with new visuals while speech continues Source item — shows that AGENTS.md was captured from VS Code Live waveform — makes audio reception immediately visible Background transformation — shows that speech is still becoming text Pending recordings — accumulates unpasted speech into one session
1 2 3 4 5
1 / 5
Captured items — the stack grows with new visuals while speech continues
Captured items — the stack grows with new visuals while speech continues
Speak, capture, and speak again; pending recordings stay together until delivery.
2 — When you say “this part,” capture the screen and mark the region. Choose Screen Capture from Quick Actions or use the capture shortcut. The rectangle enters the conversation as a clickable O1.R1-style reference, so you do not need to write a separate explanation list.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Live speech flow — shows the session state while recording continues Target app — shows the context where the conversation started Pipeline stages — hovering opens the transformed-conversation panel Transformed conversation — keeps speech, annotations, and visual context together Inline O1.R1 reference — the first area marked while speaking Other rectangle references — connect drawn image regions to the conversation Captured-image reference — the O1 name and capture identity View selector — switches between Conversation, Events, and Context Events — shows the structured context sent to the agent Conversation — shows plain human speech without annotation tokens Delivery target — shows that this conversation is headed to Codex Quick Actions — opens capture actions by click or shortcut Screen capture — adds a new screen to the annotation flow Captured-item stack — gathers items attached to the conversation Source app and title — shows that the capture came from Cockpit Representation — shows that the item will be sent as an Image and can be changed Approximate magnitude — tokens when known, otherwise kilobytes
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 / 17
Live speech flow — shows the session state while recording continues
Live speech flow — shows the session state while recording continues
Speech, target app, screen captures, and rectangle references stay in one uninterrupted session.
3 — Check which app and visual the conversation belongs to. The live flow shows the target app. Conversation opens plain speech, Events exposes the structure sent to the agent, and Context opens the combined human-readable explanation. Each captured-item row carries its source app, representation, and approximate magnitude.
4 — You do not need to paste immediately. If you start another recording, the previous unpasted recording waits. When you deliver, pending recordings merge into one session instead of making you rebuild context by hand.
5 — Find an older conversation in Dictation Browser. Search by duration, user tag, relationship, text, or a suggested filter. Select a word, pause, or waveform position in the chosen recording to play audio from there.
1 2 3 4 5 6 7 8 9 10 11 12
Conversation list — shows all Dictation recordings in one place Duration — shows the length of each raw conversation Tags — user labels added with Command-D or the context menu Relationship mark — records the link between conversations pasted together Selected conversation — shows the human's transcript in detail Relationship strip — shows other conversations connected to this recording Transformation trace — exposes a phrase removed by the custom cleanup algorithm Audio playback bar — plays from a word, pause, or waveform position Natural pause — shows a silent area in light gray Total speech duration — combined speaking time of the panel's recordings Search — finds recordings by text, tag, and capability expression Suggestions — offers matching filters as the query changes
1 2 3 4 5 6 7 8 9 10 11 12
1 / 12
Conversation list — shows all Dictation recordings in one place
Conversation list — shows all Dictation recordings in one place
Search, suggestions, duration, tags, relationships, and playback live on the same recording surface.
6 — Follow the context tree through nested visuals. The first tour moves from the assembled prompt into recording history, live Dictation, the hover preview, and Cockpit. Its other branch moves from the capture stack to the Codex delivery. Each screen opens from the context of the previous visual.
1 2 3
Opening recordings — starred voice conversations that begin the prompt Continuation recordings — the other starred conversations collected for the same task Assembled prompt — one agent packet containing the raw speech and visual context
Conversation list — shows all Dictation recordings in one place Duration — shows the length of each raw conversation Tags — user labels added with Command-D or the context menu Relationship mark — records the link between conversations pasted together Selected conversation — shows the human's transcript in detail Relationship strip — shows other conversations connected to this recording Transformation trace — exposes a phrase removed by the custom cleanup algorithm Audio playback bar — plays from a word, pause, or waveform position Natural pause — shows a silent area in light gray Total speech duration — combined speaking time of the panel's recordings Search — finds recordings by text, tag, and capability expression Suggestions — offers matching filters as the query changes
Live speech flow — shows the session state while recording continues Target app — shows the context where the conversation started Pipeline stages — hovering opens the transformed-conversation panel Transformed conversation — keeps speech, annotations, and visual context together Inline O1.R1 reference — the first area marked while speaking Other rectangle references — connect drawn image regions to the conversation Captured-image reference — the O1 name and capture identity View selector — switches between Conversation, Events, and Context Events — shows the structured context sent to the agent Conversation — shows plain human speech without annotation tokens Delivery target — shows that this conversation is headed to Codex Quick Actions — opens capture actions by click or shortcut Screen capture — adds a new screen to the annotation flow Captured-item stack — gathers items attached to the conversation Source app and title — shows that the capture came from Cockpit Representation — shows that the item will be sent as an Image and can be changed Approximate magnitude — tokens when known, otherwise kilobytes
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Live-flow summary — hovering opens its detailed preview Annotated Cockpit preview — shows the referenced image with its marked regions Files and Image selector — opens the content's source file or visual view Source context — shows that the capture came from Cockpit and optionOS
Terminal filters — reveal open Claude, Codex, and Terminal sessions Expanded agent card — gathers the session's essential information in one card Card index — shows that this is agent 333 in the complete sequence Agent identity — provider icon, session title, and primary information Goal — appears on the card when the agent has set one Terminal source — the terminal running the agent; Ghostty here Live-stream control — starts or pauses tracking the agent's changes Session activity — messages, subagent conversations, and file changes User messages — only the first and last human message Agent messages — only the first and last agent message Subagent conversations — messages exchanged with spawned subagents Edited files — number of files edited by the agent Uncommitted files — changed files that have not entered a commit Committed changes — number of changes already committed Conversation actions — experimental copy options Copy human messages only Copy editable context — messages and changed-file summaries Last meaningful activity — working area and elapsed time Fork — separates the session into a new terminal flow Compact — condenses the conversation with optionOS's own mechanism
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Captured items — the stack grows with new visuals while speech continues Source item — shows that AGENTS.md was captured from VS Code Live waveform — makes audio reception immediately visible Background transformation — shows that speech is still becoming text Pending recordings — accumulates unpasted speech into one session
Codex session — the Ghostty tab that received the conversation Transcript packet — spoken content delivered without manually typed companion text
1 2
1 2 3
1 2 3 4 5 6 7 8 9 10 11 12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 2 3 4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5
1 2
1 / 3
Opening recordings — starred voice conversations that begin the prompt
Opening recordings — starred voice conversations that begin the prompt
Five recordings → one prompt → Dictation → Cockpit or Codex. Click through the nested visual path.
Seven screens and 63 marked regions share one tour; the hover door also names the real app's hover behavior.
7 — Watch the Cockpit card while the agent works. The card shows the goal, terminal, first and last human and agent messages, subagent conversations, and file states. Copy only human messages, take editable context, fork, or use the optionOS Compact action when needed.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Terminal filters — reveal open Claude, Codex, and Terminal sessions Expanded agent card — gathers the session's essential information in one card Card index — shows that this is agent 333 in the complete sequence Agent identity — provider icon, session title, and primary information Goal — appears on the card when the agent has set one Terminal source — the terminal running the agent; Ghostty here Live-stream control — starts or pauses tracking the agent's changes Session activity — messages, subagent conversations, and file changes User messages — only the first and last human message Agent messages — only the first and last agent message Subagent conversations — messages exchanged with spawned subagents Edited files — number of files edited by the agent Uncommitted files — changed files that have not entered a commit Committed changes — number of changes already committed Conversation actions — experimental copy options Copy human messages only Copy editable context — messages and changed-file summaries Last meaningful activity — working area and elapsed time Fork — separates the session into a new terminal flow Compact — condenses the conversation with optionOS's own mechanism
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1 / 20
Terminal filters — reveal open Claude, Codex, and Terminal sessions
Terminal filters — reveal open Claude, Codex, and Terminal sessions
Twenty marked regions expose the compact agent card from identity through Compact.
