# optionOS

> optionOS is a macOS app that turns spoken work sessions — speech, screenshots, screen recordings and marked screen regions — into one agent-ready task packet with intent, evidence and exact references.

## What makes optionOS different

**Core experience:** “I didn’t describe anything. I just pointed.”

optionOS records what you meant, not merely what happened. While you speak, it binds screenshots, screen recordings and marked screen regions to the exact moment and intent, then hands the agent one provenance-linked task packet.

- Sentence-level temporal grounding: a mark belongs to what you were saying when you made it.
- Intent bound to evidence: the packet keeps speech, visual source, marked region and exact reference together.
- Human-triggered capture: optionOS records what you deliberately include, not an ambient history of everything on screen.
- Beyond dictation: its output is task context for an agent, not merely cleaned text.

## What you can turn on

- **Activity Disclosure** — Shows one consistent card while OptionOS reaches a protected resource. Without it, you can't see what is running in the background.
- **Agent Continuation** — Carries the conversation losslessly into the next context: it applies the carry budget when the window fills. Without it, the agent falls back to the provider's own summary.
- **Agent Transcript** — Makes agent conversations readable in session surfaces and exports them. Without it, past conversations aren't reachable from here.
- **Annotation** — Opens screenshot, GIF, OCR, and the markup editor with their shortcuts. Without it you describe what you mean in words, and the agent can't tell exactly where you pointed.
- **Cockpit** — Gathers live agent sessions, terminal surfaces, and session history in one deck. Without it, you follow each session from its own terminal.
- **Cursor Follower** — Shows live work status next to the pointer. Without it, you return to the window to see whether work finished.
- **Dictation** — Turns what you say into text and streams it with its context; every recording stays searchable together with its images, pastes and tags. Without it your speech never becomes text, and you can't search back for the images you captured or the things you copied.
- **Agents** — See and answer the agents running on your Mac from your phone: asking, waiting and working are counted separately. Without it you only learn that an agent is waiting for you once you are back at the Mac.
- **Dictation** — Speak on your phone and attach the images; the recording leaves knowing which agent it belongs to. Without it you jot the idea down and explain it again at the Mac.
- **Mac Link** — Pairs the phone with the Mac so capture, sharing and Cockpit actions flow through the same session. Without it the phone and the Mac stay two separate worlds and you carry the context by hand.
- **Tab Gestures** — An anchored finger plus a short tap moves through tabs, history, and undo/redo. Without it, you make those moves from the keyboard.
- **Terminal Capture** — Reads the active agent session in VS Code and Ghostty and tags dictation, clipboard, and search with it. Without it you have to say by hand which recording belongs to which agent.

### Experimental — opt in from Settings

- **iPhone Link** — Connects this Mac to your iPhone for capture, sharing, voice, and Cockpit actions. Without it the phone and the Mac stay two separate worlds, and you explain the idea again once you are back at the Mac.
- **Order History** — Pulls your orders from supported shopping sites through your browser session and keeps them searchable on your machine. Without it you browse each site's order pages one by one to find what you bought and when.
- **Speaker Isolation** — Marks other people's words in your text instead of dropping them. Without it, nearby conversations blend into your text unmarked.
- **Speaker Separation** — Splits overlapping voices into separate channels. Without it, passages spoken at the same time stay tangled.
- **Text to Speech** — Reads selected text and clipboard entries aloud. Without it, you read every text yourself.

### What you get once it is on

- **App Context** — Binds the app and the window you were in to the recording as you speak. Without it you would work out afterwards which task the recording came from.
- **Search** — Find what you are looking for among the agents running right now, from your phone. Without it you would scroll the list to find the agent you want.
- **Speech-Aligned Marks** — The screen region you mark is bound to the moment you spoke about it; the recording opens both together. Without it the moment you said "this one" and the place you pointed at would drift apart in the recording.
- **Speech History** — Every recording keeps the images you captured, what you pasted and your tags beside it, searchable by kind, app, time or text. Without it you would open recordings one by one to find what you said and what you captured at that moment.
- **Targeted Delivery** — You flag a target before you speak, and the text goes straight to that agent or window. Without it you would copy the text, find the right window and paste it yourself.

## Start here

- [Give AI not only what you said, but also what you saw and which part you meant](https://optionos.app/guide/en/voice-driven-ai-development.md): The idea is simple: do not try to describe the code first; show what you see, speak what you want, and hand the context to the agent without losing the thread. optionOS connects speech, capture,…
- [The voice task flow end to end: speak, mark, merge, hand to the agent](https://optionos.app/guide/en/voice-task-flow.md): Every screen relates to the others; this page merges them into one structure. The flow below is not separate images: it is a single journey that tells what a person does from start to finish, with…
- [Do not send a visual alone; carry its marks, explanation and source relation together when they exist](https://optionos.app/guide/en/lineage-graphs.md): A lineage graph does not leave a visible UI piece as a standalone screenshot. The image, any marks that actually appear on it and their meaning, the component/source-map trace and the spoken intent…
- [Grow while awake, strengthen the system while asleep: my agentic work loop](https://optionos.app/guide/en/agentic-work-loop.md): I want to show how I actually work right now. This is not an abstract “I work with AI” story. It is the construction site in front of me at that exact moment: how long the agents have been running,…

## More guides

- [There are no active promises outside a release right now](https://optionos.app/guide/en/promises.md): Experimental features and roadmap ideas are not promises. An outcome becomes a promise only when it is explicitly included in a specific release's closed contract and has a written way to falsify it.
- [How does an AI agent find the right rule without reading all of them?](https://optionos.app/guide/en/anchor-mcp-how-it-works.md): Giving an agent every rule on every task fills its context.
- [When does an idea from a conversation become a permanent rule?](https://optionos.app/guide/en/conversation-to-rule.md): `text Raw explanation → correct reality errors → match against existing rules → find the measured gap → extend the correct existing rule → add ground + falsifier `
- [I run an autonomous agent with one prompt, rules, and context recovery](https://optionos.app/guide/en/rule-driven-autonomous-agent.md): This flow shows, on the original screens, how one prompt freezes the target, how semantic deltas and a closed file scope make the work auditable, and how the target and rules return after context…
- [I have Rules review a Forge plan and send it back through Cockpit](https://optionos.app/guide/en/multi-agent-architecture-review-loop.md): Let me walk you through the way I work; we are doing this as a guide. First of all, I have an agent: an agent I call the Forge agent, and it presents a report to me. Here it has produced a long…
- [How my agents stay on track across context changes: anchor, stepper and conversation-cap](https://optionos.app/guide/en/agent-harness-loop.md): A goal alone is not what keeps an agent on target for twelve hours. My optionOS harness preserves three relationships together:
- [Choose the reader question from the top tab and the content type from tags](https://optionos.app/guide/en/field-guide-structure.md): Field Guide top tabs are not content types; they are the reader’s big questions. Rules, bets, workflows or features appear as global tags; the tab only decides which big question the page answers.
- [Turn on Cockpit — Ghostty is automatic, VS Code is optional](https://optionos.app/guide/en/cockpit-setup.md): Cockpit is a feature of the optionOS Mac app; you turn it on in settings. Claude Code and Codex sessions running in the Ghostty terminal are captured with no extra setup — the VS Code extension is an…
- [Frequently asked questions](https://optionos.app/guide/en/faq.md): The most common questions about optionOS: working with agents, offline use, privacy and pricing. Every answer describes today's behavior; when behavior changes, this page changes with it.
- [optionOS vs Wispr Flow](https://optionos.app/guide/en/optionos-vs-wispr-flow.md): Wispr Flow is a dictation app: in its own words, "Don't type, just speak" — it turns your speech into clean, edited text in any app, and it does that well. Its output is text.
- [optionOS vs superwhisper](https://optionos.app/guide/en/optionos-vs-superwhisper.md): superwhisper is a dictation app: in its own words, "Just speak. Write faster." — it turns your speech into polished, formatted text and works offline. Its output is text.
- [Best voice input for Claude Code: how to choose](https://optionos.app/guide/en/best-voice-input-for-claude-code.md): If you want to give Claude Code or Codex work by voice, there are two classes of tools. The first is dictation apps (like Wispr Flow or superwhisper): they turn your speech into clean text that you…
- [I produce everything by speaking — these pages included](https://optionos.app/guide/en/voice-only-content-production.md): I produce this site without typing a single line. I record my own voice conversations, tag the useful ones in the panel, then have agents produce from those conversations. The marketing copy, these…
- [How I work: I speak instead of typing and mark ambiguity on the visual](https://optionos.app/guide/en/voice-built-work-method.md): The content on this page did not begin as a line-by-line written prompt. I made five voice recordings, captured screens at the important moments, and marked 63 regions while speaking. The raw…
- [Use speech, visuals, and marked regions as one task context](https://optionos.app/guide/en/voice-context-workflow.md): 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…
- [This is how I work with agents: lineage trees, rules and building by talking](https://optionos.app/guide/en/agentic-work-loop-claude.md): I want to show my own way of working. Everything on this page came out of roughly thirty minutes of live talking: I spoke while looking at my screen, copied and marked what I saw as I spoke, then…
- [From ⌥A to close: everything a voice session collects while you talk](https://optionos.app/guide/en/voice-session-walkthrough.md): This guide was produced with the very method it describes: one 12-minute spoken session, rectangles drawn while talking, and captures taken along the way. Every visual below is interactive — hover or…
- [Speak, mark the areas and turn the same visual into a clickable explanation](https://optionos.app/guide/en/interactive-visual-lesson.md): This example uses the screenshot and two marked areas sent during the conversation. Hover or click area 1 or 2; the explanation, choices and next target change inside the same card.
- [Measure whether documentation lineage reduces duplicate and orphan decisions](https://optionos.app/guide/en/prediction-log.md): Claim: documentation blocks can disappear inside large AI context if they do not carry unique anchors and lineage. Expected effect: documentation writers add not only prose, but objective node id,…
- [Connect documentation blocks with unique anchors, one authority and lineage](https://optionos.app/guide/en/context-forest.md): When an AI reads many prompts, policies, skills, guides and notes, it no longer sees one document at a time; it receives a large context forest. A small but important block can disappear when its…
- [If Fable 5 pauses for safety, do not repeat the problematic wording verbatim](https://optionos.app/guide/en/fable-5-taklit-et-keyword-safety-flag.md): Three benign phrases or contexts encountered Fable 5's safety-pause surface during normal optionOS development. The goal is not to weaken safety controls. It is to carry harmless development intent…
- [Take a screenshot for something static; record the screen for motion; label the area to show what you mean](https://optionos.app/guide/en/speak-capture-annotate-task.md): The label does not have to be a component — it can be any region on the screen.
- [Read an agent session in its Cockpit card and continue from there](https://optionos.app/guide/en/cockpit-agent-card-data-flow.md): This page shows the agent-card screenshot from What's New as the same image-card lineage, not as a bare image. The caption and the 1–15 indexed legend come from the What's New source entry; the Guide…

## What's New

- [Release notes](https://optionos.app/whats-new/en/): every shipped feature with visual evidence.
- [Cockpit cards show an agent session as individual data-flow points](https://optionos.app/whats-new/en/2026-06-29-cockpit-mac-agent-card-data-flow/): A Cockpit card now shows an agent session as separate readable data points instead of only a one-line summary. The title, target, session identity, model, host app, state, transcript action, file…
- [The optionOS ecosystem in one working surface](https://optionos.app/whats-new/en/2026-07-02-optionos-ecosystem-family/): optionOS now reads less like separate tools and more like members of the same AI work ecosystem: Dictation carries speech and capture, Clipboard/Pano manages visuals and paste flow, and Cockpit…
- [Cockpit shows usage limits and stable Ghostty cards](https://optionos.app/whats-new/en/2026-07-06-cockpit-mac-usage-limits-ghostty/): The Mac Cockpit grid now shows Codex 5h/7d usage percentages in the Usage chip. Opening it keeps the currently available Claude and Codex subscription account details on one surface. Card state…
- [v0.5: The system behind speech is now visible](https://optionos.app/whats-new/en/2026-07-11-optionos-mac-visible-background-system/): Recording, transcription, capture, paste, storage and agent flows used to run mostly in the background; only the person building the system could see what it was doing. v0.5 brings states, choices,…
- [Cockpit cards now have a stable number, a color tag, and two transcript-copy scopes](https://optionos.app/whats-new/en/2026-07-13-cockpit-mac-card-memory-transcript-copy/): Cockpit cards now make it easier to find an agent session again and choose exactly which transcript scope should travel with you.
- [Sign in to an account once on this Mac; Cockpit keeps its usage beside your other accounts](https://optionos.app/whats-new/en/2026-07-13-cockpit-mac-multi-account-usage-history/): Switching accounts no longer makes the previous usage view disappear. Cockpit remembers the Claude and Codex accounts observed on this computer and keeps their last-known 5h/7d usage together in one…
- [Screen-recording permission now leads to the right Settings page from one card](https://optionos.app/whats-new/en/2026-07-13-optionos-mac-screen-recording-handoff-guide/): When screen-recording access is needed, the correct System Settings destination and every required step remain in one guide card.
- [Stop choosing a paste format; optionOS now selects it for the target app](https://optionos.app/whats-new/en/2026-07-13-optionos-mac-smart-automatic-target-paste/): You no longer need to read a separate format card or choose a mode before pasting.
- [Speech flow and the source of captured content are now visible](https://optionos.app/whats-new/en/2026-07-13-optionos-mac-terminal-agent-voice-lanes/): optionOS now shows conversation order, the current word during playback and the source of captured content directly. The terminal agent and voice turns remain separate targets, the played word…
- [Use Capture Text to copy words on screen; use Add Note to start a spoken note](https://optionos.app/whats-new/en/2026-07-14-optionos-mac-capture-text-note-actions/): Your four frequent capture actions now share one list: Screen Capture, Record GIF, Capture Text, and Add Note.
- [Flag the field first, then paste back to it even after speaking in another app](https://optionos.app/whats-new/en/2026-07-14-optionos-mac-flagged-delivery-target/): Press ⌥⇧V over a writable field to flag it as the delivery target. You can move through other apps while speaking, then press ⌥V to paste the pending dictation into the flagged field instead of the…
- [Images prepare faster, language issues are fixed, and long conversations feel smoother](https://optionos.app/whats-new/en/2026-07-14-optionos-mac-media-language-performance/): This release improves media delivery, Turkish–English text, and long-session performance together.
- [Type the start of a word; find the right recording and every transcript match together](https://optionos.app/whats-new/en/2026-07-14-optionos-mac-voice-recording-semantic-search/): Type a beginning such as si, and optionOS suggests a matching search token such as signal. Once selected, matches are highlighted both in the recording list and inside the open transcript.
- [Select a numbered area; let the visual focus and open its explanation beside it](https://optionos.app/whats-new/en/2026-07-14-website-spoken-visual-guide/): A single Guide visual now turns the areas marked during a spoken explanation into clickable steps. Each step focuses the right region and opens its explanation beside it, so you can follow how a…
- [Open Shortcuts from Settings and see when each one works from its category](https://optionos.app/whats-new/en/2026-07-15-optionos-mac-contextual-shortcuts/): Shortcuts now live on one Settings surface. Each row shows the action, assigned keys, and enabled state, while its category tells you whether it works at all times, during a conversation, or in tools…
- [Open a guide on its own page; switch language or share the exact step](https://optionos.app/whats-new/en/2026-07-15-website-guide-pages-shareable/): Website v0.1 is live. Every guide now has a directly accessible page in Turkish and English. The language control moves to the sibling page for the same guide.
- [Open the Settings tutorial from the graduation-cap button and move with the arrow keys or Enter](https://optionos.app/whats-new/en/2026-07-16-optionos-mac-settings-tutorial/): The graduation-cap button in the optionOS Settings toolbar now opens the Settings tour. Its panel shows all three targets; after selecting Start, move between steps with ←, →, or Enter.
- [Open an agent link; jump to the exact session and follow the work in Cockpit](https://optionos.app/whats-new/en/2026-07-19-cockpit-mac-terminals-wide-agent-card/): Click an https://app.optionos.app/agent?provider={provider}&session={sessionId} link shown in a conversation and optionOS opens the exact agent session: it focuses a live agent and asks for your…
- [Watch your speech live, see its tags, step inside the image you captured](https://optionos.app/whats-new/en/2026-07-19-optionos-mac-dictation-live-context-stack/): Speech is now visible end to end on the Dictation side: the live bar watches the speech and its context, the transformed panel shows the rectangles you drew and the images with their tags, and…
- [Speak, mark, and follow the agent: Dictation context now reaches Cockpit as one flow](https://optionos.app/whats-new/en/2026-07-19-optionos-mac-voice-context-workflow/): Optionalist Dictation now does more than turn audio into text. It keeps the target app, captured visuals, drawn regions, pending recordings, and the context delivered to the agent visible inside one…
- [Dictation search now responds as you type; large result sets no longer stall the panel](https://optionos.app/whats-new/en/2026-07-25-optionos-mac-dictation-panel-search-performance/): Search results refresh with each new part of the query. Search, list selection, and the detail surface remain usable with large result sets.
- [A new scene now shows the whole view before focusing the target](https://optionos.app/whats-new/en/2026-07-25-website-context-first-interactive-tours/): Open an interactive visual on the website and its numbered targets remain on the image while the scene and step count and explanation stay in the same panel.

## Home

- [optionos.app](https://optionos.app/): product overview and early access.

## Products

- [Cockpit](https://optionos.app/cockpit/)
- [Tab Gestures](https://optionos.app/tab-gestures/)

## Legal (English)

- [Data deletion](https://optionos.app/legal/en/data-deletion/)
- [EULA](https://optionos.app/legal/en/eula/)
- [Privacy](https://optionos.app/legal/en/privacy/)
- [Terms](https://optionos.app/legal/en/terms/)
- [Brand use](https://optionos.app/brand/)

## Full content

- [llms-full.txt](https://optionos.app/llms-full.txt): every English guide as full markdown in one file.

## Languages

- [Türkçe](https://optionos.app/tr/llms.txt)