№ 02Learn With Darin · Field Guide
Claude App: a practitioner’s field guide.
The same model behind a different door. A practitioner's guide to Claude on Mac, Windows, iOS, and Android: what each surface gives you, where each one quietly takes something away, and the workflows that hold up across all four.
The five surfaces, one model
"The Claude app" isn't a single thing. Anthropic ships five front doors to the same underlying model, and most of the friction people hit comes from assuming the door is the room. The doors are:
- Web at
claude.ai: the canonical surface. Everything else lags this by some amount, except where the OS gives them an advantage. - Mac desktop app, a thin Electron-style wrapper over the web with menu-bar integration, a global hotkey, file system access via Computer Use, and (since April 2026) Claude Cowork.
- Windows desktop app, the same wrapper for Windows, with the same Cowork capabilities and a different system-tray idiom.
- iOS app, a native iPhone/iPad client with Share Sheet, App Intents, Siri Shortcuts, widgets, and a few Apple-only capabilities.
- Android app, a native Android client with system share intents, widgets, and Assistant handoff.
Inside all five, you're talking to the same family of models. As of May 2026 that's Claude Sonnet 4.6 by default, with Opus 4.7 (1M context) available on Pro and above when you opt into "extended thinking" mode. Free is pinned to Sonnet 4.5. The model picker is the same picker. What differs is which models the picker offers, which is a function of your plan, not the surface.
The piece most people get wrong is treating "Claude on my phone" and "Claude on my laptop" as different products. They aren't. Conversations sync. Projects sync. Memory syncs. Artifacts you open on desktop you can hand back to mobile mid-thread. The only thing that doesn't sync is voice-mode audio history, and that's deliberate.
What this guide is going to do is walk through the four app surfaces (skipping web, which is just the default), call out where each one rewards or punishes you, and end with the limits, etiquette, and troubleshooting playbook that applies across all of them.
Plans, briefly
You can't talk about app capabilities without talking about plans, because most of the "this doesn't work for me" reports trace back to plan ceilings rather than app bugs. Here's the May 2026 picture, simplified:
| Plan | Cost | Default model | What you get |
|---|---|---|---|
| Free | $0 | Sonnet 4.5 | Web, mobile, desktop. Text + image + code. Web search. File uploads. Artifacts. Roughly 30–100 messages per 5-hour window. No Projects. |
| Pro | $20/mo (or $17/mo annual) | Sonnet 4.6 + Opus on demand | Adds Projects, code execution, Google Workspace, remote MCP connectors, Claude Code, "extended thinking" Opus access. ~5× Free's headroom. |
| Max | $100 or $200/mo | Same as Pro | 5× or 20× Pro's usage ceiling, priority on new features, no extra capabilities. Just more headroom for power users. |
| Team | $25/seat (Std) or $125/seat (Premium) | Same as Pro | Org-level admin, shared Projects, billing centralization. Premium adds Claude Code. 5-seat minimum, 150-seat ceiling. |
| Enterprise | Sales | Same as Pro | 500K context (vs 200K), HIPAA-ready, SSO, SCIM, audit logs, custom retention. The plan you upgrade to once Legal asks "where does our data go." |
One thing the table doesn't capture: usage limits are 5-hour rolling windows, not daily. If you have a heavy hour at 2pm, you'll have less headroom at 5pm but full headroom by 7pm. The mobile and desktop apps don't tell you where you are in that window; only the web app's settings page does. That asymmetry catches people out, especially on long voice-mode walks.
Mac vs Windows
The desktop apps look almost identical. They are not identical. The differences come from what each OS lets a client app do, and they cluster around: launch behavior, file access, voice, and the keyboard.
Install & update model
Both ship as native installers. The Mac app self-updates silently in the background; you'll notice when a new feature shows up overnight. The Windows app prompts you on launch when an update is available, and you have to click through. If you're on Windows and noticing your friend's Claude app does something yours doesn't, check the version number first.
Distribution: Mac is .dmg from claude.com/download; Windows is .exe from the same place. Neither is in the App Store or the Microsoft Store, which means corporate-managed devices may need IT involvement.
Where it lives
macOS
- Lives in the menu bar as well as the Dock. The menu-bar icon is the fastest path to a new chat.
- Default global hotkey: ⌘+Shift+Space opens a quick prompt overlay.
- Window state restores cleanly across reboots and logouts; conversations resume where you left them.
- "Hide" (⌘+H) keeps the app warm; quitting fully (⌘+Q) ends voice sessions.
- Spotlight indexes the app name but not conversation contents.
Windows
- Lives in the system tray (right side of the taskbar). The tray icon is the equivalent of macOS's menu-bar icon.
- Default global hotkey: Alt+Space opens the same quick prompt overlay.
- Window state is less reliable across full reboots; expect to relaunch into a fresh window occasionally.
- Closing the window minimizes to tray; "Exit" from the tray menu actually quits.
- Windows Search does not index Claude content; the Start menu only finds the app itself.
The hotkey difference is the one that matters in practice. ⌘+Shift+Space on Mac collides with nothing common; Alt+Space on Windows collides with the legacy "system menu" command on focused windows. If you're a heavy keyboard user on Windows, change the hotkey first thing in Settings → General → Hotkey.
File pickers and Computer Use
Both apps can read files via drag-and-drop, the attachment button, and (on Pro and above) Computer Use, which lets Claude operate your file system and other applications under permission.
macOS
- Drag-and-drop accepts almost anything: PDFs, images, code,
.docx,.xlsx,.csv, plain text up to plan limits. - Sandboxed access to Documents, Desktop, and Downloads by default; Full Disk Access requires a TCC prompt and System Settings approval.
- Computer Use respects macOS accessibility permissions, and you'll be prompted on first use of UI automation per app.
- Watch Apple's per-app permission dialogs for "Files and Folders." Declining them silently breaks file actions later.
Windows
- Drag-and-drop works the same, with one caveat: drops from OneDrive on-demand folders that haven't been hydrated will fail silently.
- No equivalent to TCC; file access is broader by default, which is both easier and a bit scarier.
- Computer Use uses UI Automation; some legacy Win32 apps with custom rendering (think CAD or older finance terminals) are partially or fully invisible.
- Antivirus and DLP agents occasionally flag the file-access behavior. If actions disappear from menus, check your endpoint logs.
Voice mode
Voice mode is one feature that's reached parity across all four apps in early 2026, but the desktop builds expose it differently. Both Mac and Windows put voice mode behind the microphone icon next to the input box; both default to push-to-talk and let you switch to continuous listening from the in-call menu. Five voices are available (Buttery, Airy, Mellow, Glassy, Rounded), and the choice persists per device, not per account.
The differences:
- Mac uses the system microphone the OS picks unless you override in System Settings → Sound. AirPods Pro switching is handled at the OS level and works reliably.
- Windows respects the default communication device, which can lag the default playback device by a beat. If voice cuts in but Claude can't hear you, your default microphone is wrong. Fix it in Settings → System → Sound.
- Mac shows a menu-bar indicator while voice mode is hot; Windows shows a tray badge and a tooltip. The Windows indicator is easier to miss.
Cowork and the desktop apps
Since April 9, 2026, both desktop apps include access to Claude Cowork for users on paid plans. Cowork is its own product surface (see the Cowork guide for the full treatment), but it shares a window with the regular Claude app. Cowork features (Connectors, Plugins, Scheduled Tasks, Dispatch, Computer Use) appear as a new top-level mode in the sidebar; switching modes does not switch your conversation history.
The practical implication for the consumer experience: on desktop, you're never more than a click away from giving Claude permission to operate your machine. That's powerful and a permission posture worth thinking about.
iOS vs Android
Mobile is where the per-platform differences get pronounced. Both apps share a code base for the chat surface itself, but the OS-integration features diverge sharply because Apple and Google give app developers different primitives to work with.
The basic chat experience
From inside a chat, the two apps are nearly indistinguishable. Same model, same Projects sync, same file uploads, same Artifacts rendering, same voice mode with the same five voices. Both run identically on phone and tablet form factors. If you're just having conversations, pick whichever phone is in your pocket.
The fork shows up the moment you try to get content into Claude or do something with what Claude returned.
Share sheets & getting content in
iOS
- Standard Share Sheet integration: anything sharable from any app (webpages, photos, PDFs, selected text) has a "Share to Claude" target.
- "Summarize with Claude" appears as an additional Action in the Share Sheet for selected text, useful for skimming long emails or articles in place.
- Multi-file selections from Files.app share into a single chat, preserving filenames and types.
- Shortcuts.app exposes Claude actions: "Ask Claude," "Summarize Text with Claude," "Transcribe with Claude," each with input and output that can be wired into automations.
- App Intents: third-party apps can hand work to Claude programmatically. Few do this in May 2026, but the surface area is there.
Android
- Android share intents are honored: long-press text or hit Share in any app and Claude appears as a target.
- No equivalent of "Summarize with Claude" inline action; you share into a Claude chat and prompt from there.
- Multi-file shares are flaky on Android 13 and below; reliable on 14+.
- Tasker integration is community-driven (no first-party Tasker plugin from Anthropic). It works because of standard intents, not because there's a Claude-specific Tasker action.
- Quick Settings tile available. Drop an "Ask Claude" tile into your pull-down for one-tap voice or chat.
The iOS Shortcuts story is meaningfully better than anything on Android right now. If you're an automation-heavy user, that asymmetry is real. On the other hand, the Android Quick Settings tile beats anything iOS offers from a "fastest possible voice prompt" angle, with fewer taps from any app, including the lock screen.
Voice mode
Voice mode is the feature that benefits most from being on a phone. Both apps support continuous voice mode with interruption handling: you can talk over Claude, and it'll yield. Both apps offer the same five voices.
Differences worth knowing:
- iOS supports voice mode with screen off (background audio capability), which means you can keep walking and talking without the screen on. The phone's silent switch is honored: flip it and Claude responds in text only.
- iOS integrates with Siri Shortcuts for voice triggering: "Hey Siri, ask Claude…" works if you've set up the shortcut. The handoff is jankier than native Siri because Apple gates intents, but it works.
- Android requires the screen to remain on for continuous voice mode unless you've granted "Display over other apps," a permission that, fairly, freaks people out. Without it, voice mode pauses when the screen times out.
- Android's integration with Google Assistant is read-only: Assistant can hand off to Claude via the share intent, but Claude can't be the system-default assistant.
Camera, vision, and photos
Both apps let you tap a camera icon to take a photo and analyze it without leaving the app. Both let you attach existing photos via the standard photo picker. Both support multiple images per turn (up to plan limits, usually five).
Where they diverge: iOS uses the modern PHPicker, which means Claude only sees the photos you explicitly select. Your full library is never enumerated, even with permission. Android's photo picker is similar on Android 13+, but on older versions you grant broader Photos permission and Claude trusts you to pick wisely. The privacy posture is materially different.
Widgets
iOS
- Home Screen widget with one-tap buttons: New Chat, Voice, Camera.
- Lock Screen widget (small + circular) for one-tap voice. Hold the phone, speak, done.
- Today View widget with the same triple-button layout.
- StandBy mode-friendly clock-style widget that surfaces "Claude is ready" with a tap-to-talk affordance.
Android
- Home Screen widget with the same New Chat / Voice / Camera triple.
- No lock-screen widget; Android's lock-screen surface is more constrained than iOS 16+.
- Quick Settings tile (one-tap from any app, including from inside another full-screen app; the iPhone has no equivalent of this surface).
- App-shortcut menu on long-press of the launcher icon: New Chat, Voice, last conversation.
Net: iOS wins on lock-screen and StandBy surfaces; Android wins on Quick Settings. Pick whichever pattern fits your physical interaction with the phone.
Background limits and "where did my chat go"
Mobile OSes are aggressive about backgrounded apps, and Claude's behavior diverges:
- iOS rarely kills the Claude app outright unless memory pressure is extreme; conversations resume in place. Voice mode is treated as background audio and survives lock.
- Android's aggressive battery managers (especially Samsung One UI's "Sleeping apps" and Xiaomi's MIUI background killers) will silently kill the Claude process. Add Claude to your battery-optimization allowlist if you use voice mode regularly. Otherwise you'll get cut off mid-sentence.
Capability matrix
Most features are everywhere, but not all of them. Bookmark this table; it answers most "why doesn't X work on my…" questions.
| Capability | Web | Mac | Windows | iOS | Android |
|---|---|---|---|---|---|
| Text chat | ✓ | ✓ | ✓ | ✓ | ✓ |
| File upload (PDF, image, code) | ✓ | ✓ | ✓ | ✓ | ✓ |
| Artifacts (rendered) | ✓ | ✓ | ✓ | ✓ | ✓ |
| Projects (Pro+) | ✓ | ✓ | ✓ | ✓ | ✓ |
| Memory | ✓ | ✓ | ✓ | ✓ | ✓ |
| Web search | ✓ | ✓ | ✓ | ✓ | ✓ |
| Voice mode (5 voices) | ✓ | ✓ | ✓ | ✓ | ✓ |
| Code execution sandbox (Pro+) | ✓ | ✓ | ✓ | ✓ | ✓ |
| Computer Use (Pro+, via Cowork) | – | ✓ | ✓ | – | – |
| Cowork: Connectors, Schedules, Dispatch (paid) | partial | ✓ | ✓ | partial | partial |
| Remote MCP connectors (Pro+) | ✓ | ✓ | ✓ | ✓ | ✓ |
| Google Workspace integration (Pro+) | ✓ | ✓ | ✓ | ✓ | ✓ |
| Share Sheet / Share Intent | – | – | – | ✓ | ✓ |
| Siri Shortcuts / App Intents | – | – | – | ✓ | – |
| Quick Settings tile | – | – | – | – | ✓ |
| Lock Screen widget | – | – | – | ✓ | – |
| Global hotkey overlay | – | ✓ | ✓ | – | – |
| System-tray / menu-bar entry | – | ✓ | ✓ | – | – |
"Partial" on the mobile Cowork rows means: you can see a Cowork-dispatched task's status, but you can't kick one off from mobile yet. As of May 2026, Cowork's authoring surface is desktop-only.
Limits, and how they actually manifest
The headline numbers are easy to recite ("200K context, 5-hour windows, ~30-100 messages on Free"). What's useful is knowing how those limits show up in practice across the apps.
Context window
Standard accounts get 200K tokens. Enterprise gets 500K. Long conversations don't gracefully truncate. Instead, Claude starts losing detail from the middle of the conversation as you near the ceiling. Symptoms: it forgets a stipulation you made earlier, or starts repeating itself, or gives a worse answer to a question it nailed two turns ago.
The apps don't surface where you are in the context budget. There's no "65% used" bar. Watch for the symptoms, not a number. When you see them, start a new conversation and either re-paste the load-bearing parts or use a Project to share context.
Attachment ceilings
- Per-message: typically 5 files, total <100MB.
- Per-PDF: ~32MB and 100 pages on most plans; Enterprise raises both.
- Image limits: ~20MB per image, max 5 per turn. Mobile compresses uploads, so if you're sharing a screenshot of small text, prefer desktop.
- Project knowledge files (Pro+): aggregate budget separate from per-message uploads; Pro is roughly 100K tokens, Team raises it, Enterprise lets you set it.
Rate limits and the 5-hour window
Anthropic publishes rate caps as messages per 5-hour rolling window. The exact ceiling moves with model selection (Opus burns more budget than Sonnet) and demand. Free is the most variable; Pro is steadier; Max is rarely an issue in practice.
What you'll experience when you hit the cap:
- The web app shows a yellow banner with a countdown.
- The desktop apps show the same banner but truncated; on Windows it sometimes appears for one frame and disappears.
- The mobile apps show a modal dialog with the wait time.
- Voice mode degrades silently: it doesn't end the call, it transparently routes you to a smaller model, and the responses get noticeably less coherent. This is one of the most frustrating failure modes. The only fix is to end the voice call, send one text message to confirm the model fallback, and decide whether to wait or downgrade your expectations.
What's not on Free
- Projects (the entire feature, not just a quota).
- Code execution sandbox.
- Remote MCP connectors.
- Google Workspace integration.
- Computer Use / Cowork.
- Opus and "extended thinking" models.
Free is a functional chat product. It's not a productivity platform.
Best practices, by surface
The apps reward different patterns. Here's what holds up across months of daily use.
Use Projects for anything that survives a single conversation.
If you'll come back to the topic next week, it belongs in a Project. Your style guide, your résumé draft, the spec for the side project, the recurring research topic. The cost of starting a Project is near zero; the cost of not using one is rebuilding context every conversation. Available on Pro and above, syncs across all five apps.
Treat Memory as a default-off feature you opt into per-Project.
Memory is global by default on Pro+, but the surprises it produces (recalling a detail from a different topic in a moment when it's irrelevant) outweigh the convenience. Turn off global Memory in Settings; let each Project's own context do the remembering. Memory is at its best when scoped tight.
Voice mode is a thinking tool, not a typing replacement.
It's at its best when you're walking, driving, or pacing, anywhere talking is your actual mode and a screen would be a distraction. It's at its worst for tasks that need precise output (code, structured documents, anything you'll copy-paste). Don't dictate prompts you'd write more carefully; use voice for the part of the work that's exploratory.
Use the desktop app when you're touching local files.
The web doesn't have your filesystem. The mobile apps have a sliver. The desktop apps have the whole disk (with permission), via Computer Use. If your task involves "look at this folder" or "open that document and pull a chart out," it's a desktop task. Don't fight the tool.
Keep Free as a fallback, not a primary.
If you only use Claude occasionally, Free is fine. If you use it daily, the per-incident frustration of the cap costs more than $20/month. Pro is the sweet spot for nearly everyone; Max is for people who notice they're hitting Pro's ceiling more than once a week.
On mobile, use the share sheet, not copy-paste.
Both iOS and Android share sheets preserve metadata that copy-paste loses: source URL, file type, multi-file context. If you're handing a webpage or a PDF to Claude, share it; don't copy text out of it.
If Claude is stuck, end the chat. Don't keep trying.
A degraded conversation rarely recovers. If Claude gets stuck repeating itself, missing context, or producing weaker answers, start a new chat. Conversations are cheap. Re-paste the load-bearing context if needed. The minutes you spend rescuing a bad thread are minutes you'd spend better starting fresh.
Pin the apps you actually use; don't sync them all.
Many people install all four because they can. Most use one daily, one occasionally, two never. Figure out which two are yours and uninstall the rest. Fewer surfaces means fewer half-versions of you to keep straight (and fewer push notifications).
A note on privacy posture
This isn't a guide to Anthropic's privacy policy (that's their job to publish and yours to read), but a practitioner has to know what each app surface is sending where.
- By default, on Pro and below, Anthropic does not train on your conversations. There's an opt-in toggle for "help improve Claude" that does. It is off by default. Check it in Settings → Privacy.
- Team and Enterprise plans have stricter contractual posture by default; admins can centrally disable any opt-in.
- Voice mode does not store audio, only the transcript. Conversations transcribed from voice look the same as typed.
- Memory's contents are visible to you; you can edit and delete entries. They're scoped to your account, not your team.
- Computer Use sessions log the actions Claude took, not the screen content. The action log is in Settings → Activity.
The most underrated control is Settings → Data Controls → Delete account data. It's a hammer, but it works as advertised. If you've ever shared something into a chat that you regret, you can purge it. (On Enterprise, the admin holds the same hammer with finer granularity.)
Troubleshooting playbook
Roughly in the order you'll meet them.
Three suspects, in the order to check:
- You hit the rate cap. Check the web app's banner at
claude.ai; desktop and mobile sometimes hide it. The cap message tells you when capacity returns. - Network blip. Voice mode is especially sensitive to micro-disconnects. End the call, refresh, retry.
- Conversation context corrupted by a malformed attachment. Symptom: you can chat about other things, but this conversation refuses to respond. Start a new chat.
Projects don't actually disappear; they show up under a different filter than you remember. Check:
- The Projects sidebar's filter chip. It might be set to "Recent" not "All."
- Whether you're on the right account (most "lost Project" reports are someone signed into a personal account expecting to see a Team Project).
- Sync lag, especially on Android after coming off battery-saver. Force-quit and reopen the app.
- Microphone permission revoked? Check OS Settings → Claude → Microphone.
- Bluetooth headset hijacking input? Disconnect briefly and try with the phone's mic to isolate.
- On Android, battery optimizer killed the background audio process. Allowlist Claude in battery settings.
- If voice triggers but Claude can't hear you, your default communication microphone is wrong. On Windows, this is in Settings → System → Sound → Input.
Sync lag, almost always. Conversations are eventually consistent. Pull-to-refresh on mobile; force-relaunch on desktop. If a conversation seems split-brained between two devices, the device that started the latest message wins; let it complete and the other will catch up.
Known issue with the Windows clipboard's HTML format. Save the screenshot to disk first and drag it in, or paste into Paint and copy the bitmap back out. Claude accepts the bitmap form but rejects the HTML form Windows defaults to.
Computer Use will refuse, by design, to do anything that:
- Touches credentials it didn't see you type.
- Operates in a domain you haven't explicitly listed in the permission scope.
- Looks like exfiltration to it (uploading a folder of files to an external service it doesn't recognize).
- Triggers an OS-level security prompt it can't see (e.g., a Touch ID prompt on Mac).
Most "it refused" cases are the third bullet. If you're trying to upload data to a service Claude doesn't have a connector for, give it explicit context about why and what you're sending; that flips the heuristic.
Edit the message at the point you want to fork. The web app surfaces an "edit" pencil on hover; the mobile apps put it in a long-press menu; the desktop apps mirror web. Editing creates a branch; the original is still there, accessible via the branch arrow above the input. This is the most underused feature in the apps. Most people delete and start over instead of branching, and lose context they could have salvaged.
Closing thought
The single most useful frame for thinking about Claude across surfaces is: the model is the same; the integration is what differs. Mac gives you a menu bar; Windows gives you a tray; iOS gives you the Share Sheet; Android gives you Quick Settings; the web gives you the canonical surface where every feature lands first. None of them is "the real Claude." They're all real, with different ergonomic bargains.
Pick the surface that matches your physical relationship with the device, and let conversations flow across all of them. — TWD
If you're new to Claude, start on the web for a week to learn the affordances. Then add the surface(s) where you spend the most ambient time. Voice mode on a phone walk; Cowork on the laptop you actually work on; the mobile share sheet for whatever you're consuming on the go. The trap is treating Claude like an app you "switch to." Its strength shows up when it's the layer underneath whatever else you're doing.
And if any of this is out of date by the time you read it: claude.com/release-notes is where Anthropic posts changes, and the in-app "What's new" on the web is a faithful echo of that page. Both desktop apps lag a few days; mobile lags a few weeks.