Let a closed terminal take its own status line with it

The strip along the bottom is one line for the whole window, and connecting
writes to it, so "Connected to prod-db." outlived the tab it was about: close
prod-db, look at the terminal beside it, and the bar was still reporting on a
session that is no longer there.

A connection's own lines are now owned by the attempt that wrote them, and
closing a tab clears the line only when that is what the bar is holding.
Ownership rather than a blanket clear on close, because the alternative loses
the sentences this bar exists for — a failed save or a refused sync is what the
design deleted the bar's other four fields to make room for, and closing a
terminal is no answer to one.

.github/workflows/ci.yml

@@ -291,6 +291,7 @@ jobs:
       # so it is not done either. What reaches users is built, installed and walked through Phase 16
       # of docs/manual-checks.md by a person first.
       - name: package the windows desktop client
+        id: winpack
         if: github.event_name != 'pull_request'
         run: |
           set -euo pipefail
@@ -374,6 +375,93 @@ jobs:
           ls -la "$releases"
           echo "Packaged DodoSSH $packVersion for win-x64, from a build MinVer calls $version."
 
+          # Handed to the macOS step below rather than worked out again there. The floor logic above
+          # is thirty lines of reasoning about MinVer's pre-first-tag answer, and a second copy of it
+          # is a second thing to keep in step — while two desktop packages built from one commit
+          # carrying different version numbers is precisely the confusion this file spends that
+          # reasoning to avoid.
+          echo "packVersion=$packVersion" >> "$GITHUB_OUTPUT"
+
+      # ◆ AND THE macOS BUNDLE IS BUILT HERE, ON LINUX, AND IS ALSO THROWN AWAY.
+      #
+      # Same argument as the Windows step above, one platform along: the failures a release is most
+      # exposed to are the ones only the packager finds, and the person who would otherwise find them
+      # is the one midway through a release on the one Mac that can cut one.
+      #
+      # What this catches that the Windows step cannot: the osx-arm64 restore graph. A native package
+      # that resolves for win-x64 and has no osx-arm64 asset — libsodium and SkiaSharp both ship per
+      # RID — fails here, on every main build, rather than at the first `dotnet publish` of a release
+      # nobody can retry without a Mac.
+      #
+      # ◆ bundle, NOT pack, AND THE DIFFERENCE IS NOT A CHOICE.
+      #
+      # `vpk [osx]` cross-compiling from a non-Mac offers exactly one packaging verb: bundle, which
+      # builds the .app. There is no `[osx] pack` off a Mac, and that is correct rather than a gap —
+      # pack signs with codesign, submits to Apple with notarytool and staples the ticket, all of
+      # which is Apple tooling that exists on no other platform. So this proves the bundle and stops
+      # where the platform does.
+      #
+      # No --plist and no --icon either, deliberately. Both are proved by scripts/release-macos.sh on
+      # the machine that can also check the result; passing a rendered plist here would mean copying
+      # the substitution out of that script to no end, since nothing looks at what this produces.
+      #
+      # ◆ NOTHING IS UPLOADED, FOR THE REASON THE WINDOWS STEP GIVES.
+      #
+      # RUNNER_TEMP, dying with the job. ADR 0013 rule 3 puts the capability to ship somebody a build
+      # on a machine which is not a runner, and an unsigned .app is additionally something no Mac
+      # would open — so publishing it would be handing out a file whose only possible use is confusion.
+      - name: publish and bundle the macos desktop client
+        if: github.event_name != 'pull_request'
+        run: |
+          set -euo pipefail
+
+          # RestoreLockedMode=false for the RID, exactly as the win-x64 publish above does — see the
+          # long note there for why the committed lock files are deliberately RID-free. This runner's
+          # checkout is thrown away, so the lock files it rewrites go nowhere.
+          dotnet publish src/DodoSSH.Client.App/DodoSSH.Client.App.csproj \
+            --configuration Release --runtime osx-arm64 --self-contained true \
+            -p:RestoreLockedMode=false \
+            --output "$RUNNER_TEMP/osx-arm64"
+
+          # The apphost has no extension on macOS, so this is `DodoSSH` and not `DodoSSH.exe`. Named
+          # rather than globbed, because a publish that produced no apphost at all would otherwise
+          # bundle happily and produce an .app that launches nothing.
+          if [ ! -s "$RUNNER_TEMP/osx-arm64/DodoSSH" ]; then
+            echo "The osx-arm64 publish produced no apphost." >&2
+            ls -la "$RUNNER_TEMP/osx-arm64" >&2 || true
+            exit 1
+          fi
+
+          bundles="$RUNNER_TEMP/osx-bundle"
+
+          # The quotes around [osx] are load-bearing, exactly as they are on '[win]' above: unquoted,
+          # the shell reads it as a glob matching any one of o, s and x.
+          dotnet vpk '[osx]' bundle \
+            --skip-updates \
+            --packId DodoSSH.Desktop \
+            --packVersion '${{ steps.winpack.outputs.packVersion }}' \
+            --packDir "$RUNNER_TEMP/osx-arm64" \
+            --packTitle DodoSSH \
+            --packAuthors DodoTech \
+            --mainExe DodoSSH \
+            --bundleId dev.dodotech.dodossh \
+            --runtime osx-arm64 \
+            --channel osx \
+            --outputDir "$bundles"
+
+          # Asked for rather than inferred from an exit code, for the reason the Windows step gives.
+          # The Info.plist is the specific thing worth naming: a bundle missing it is a directory
+          # macOS will not treat as an application at all, and it is the one part of the .app that
+          # vpk composes rather than copies.
+          app="$bundles/DodoSSH.Desktop.app"
+          if [ ! -s "$app/Contents/Info.plist" ]; then
+            echo "vpk reported success and there is no Info.plist at $app/Contents/Info.plist." >&2
+            find "$bundles" -maxdepth 3 >&2 || true
+            exit 1
+          fi
+
+          echo "Bundled DodoSSH ${{ steps.winpack.outputs.packVersion }} for osx-arm64."
+
       # This includes the end-to-end suite, which starts PostgreSQL, Keycloak and an OpenSSH
       # server through Testcontainers and runs the API as a child process — so it needs a
       # Docker daemon and gets one here. That is why the tests run on ubuntu rather than

README.md

@@ -80,6 +80,8 @@ docs/platform-flags.md     what differs off Windows, and the gotchas that have c
 docs/manual-checks.md      what no test can reach, and what to look for when checking by hand
 docs/android-port.md       the Android head: what was decided, what is built, what is left
 scripts/                   release-windows.ps1 — builds, packs and publishes the Windows client
+                           release-macos.sh   — the same, signed and notarized, on a Mac
+build/macos/               the entitlements and Info.plist template the macOS bundle is built from
 ```
 
 Everything under `src/DodoSSH.Client.*` except the two heads and `Shell` is deliberately free of Avalonia.
@@ -152,8 +154,32 @@ reinstalling asks for your passphrase rather than starting over. Use **Sign out*
 you want the machine to genuinely forget everything — an uninstall is not a sign-out, and does not withdraw
 this machine's device key from your account.
 
-Cutting a release is `scripts/release-windows.ps1`, run by a person on a Windows machine. Deliberately not a
-CI job; ADR 0013 decision 3 explains why, and it is not only that the runners are Linux.
+## Installing on macOS
+
+A `.pkg` on the same release page, for Apple Silicon. Everything above about where a client may come from,
+about the update check and about uninstalling applies unchanged; what differs is worth three short
+paragraphs.
+
+**It is signed and notarized, so there is no warning to click past.** That is not generosity — macOS refuses
+to open an un-notarized download outright rather than warning about it, so unlike the Windows build there
+was never an unsigned option. If you *do* see "cannot be opened because Apple cannot check it for malicious
+software", the file did not come from the project's release page, and that is worth taking literally.
+
+**Apple Silicon only for now.** An Intel package is a small amount of work and no one here has an Intel Mac
+to check it on, and this project does not ship desktop builds nobody has run — see
+[docs/manual-checks.md](docs/manual-checks.md). Under Rosetta the arm64 build will not run; there is no
+graceful version of that, and the honest answer is that the platform is not covered yet.
+
+**Touch ID can stand in for your passphrase**, on a Mac with a Secure Enclave. The key that unwraps your
+device key is generated inside the enclave and never leaves it, and the enclave — not DodoSSH — is what
+requires your fingerprint or login password before it will use it. Cancel the prompt and you get the
+passphrase screen, always. The application lives at `/Applications/DodoSSH.Desktop.app` and your vault cache
+at `~/Library/Application Support/DodoSSH`, which are deliberately two different places so that removing the
+first never touches the second.
+
+Cutting a release is `scripts/release-windows.ps1`, run by a person on a Windows machine, and
+`scripts/release-macos.sh` on a Mac. Deliberately not a CI job; ADR 0013 decision 3 explains why, and it is
+not only that the runners are Linux.
 
 ### The nightly desktop build
 
@@ -886,8 +912,18 @@ keychain plus a terminal — and the spike that gates all of it.
   [ADR 0013](docs/adr/0013-desktop-distribution-and-updates.md), and
   [Installing on Windows](#installing-on-windows) for what a user sees.
 
-  Still to do here: signing (the first release is unsigned, and the trigger for buying a certificate is the
-  first release aimed at strangers), and macOS and Linux packaging.
+  **The macOS half is built on the same machinery**, and signed from the start because Gatekeeper leaves no
+  choice: `scripts/release-macos.sh` publishes, signs every native library, notarizes with Apple and staples
+  the ticket before it will hand anything over, and refuses to upload until a person has installed it. The
+  device key is held in the Secure Enclave behind Touch ID. CI publishes `osx-arm64` and builds the `.app`
+  on every main build to prove it still packages, and uploads nothing. See
+  [ADR 0013](docs/adr/0013-desktop-distribution-and-updates.md) decision 10 and
+  [Installing on macOS](#installing-on-macos).
+
+  Still to do here: Windows signing (the first Windows release is unsigned, and the trigger for buying a
+  certificate is the first release aimed at strangers), macOS on Intel, Linux packaging, and Phase 18 of the
+  manual checks — the macOS build has never actually run, because there is no macOS runner in CI and
+  everything above is verified only as far as the bundle.
 - **M5 — multi-provider OIDC**, identity key rotation, per-item content keys.
 
 ## Licence

build/macos/DodoSSH.entitlements

@@ -0,0 +1,67 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!--
+  What the hardened runtime has to be asked to relax before a .NET application will run under it.
+
+  The hardened runtime is not optional: notarization refuses a Developer ID submission without it,
+  and Gatekeeper refuses an un-notarized download. So every entitlement below is the price of being
+  distributable at all, and each one is a hole in a wall that is otherwise worth having. They are
+  listed one at a time, with what breaks without each, because the temptation when notarization
+  fails at eleven at night is to paste in a longer list from somewhere and stop thinking.
+
+  ◆ WHAT IS DELIBERATELY NOT HERE.
+
+  com.apple.security.app-sandbox. Developer ID distribution outside the App Store does not require
+  the sandbox, and turning it on would break the product outright: the terminal's data plane is a
+  loopback WebSocket (see DodoSSH.Client.Terminal/TerminalDataPlane.cs), and a sandboxed process
+  needs com.apple.security.network.server to listen at all, plus network.client to reach any host
+  the user asks for. This is the same shape of decision as ruling out MSIX on Windows, which was
+  ruled out for the same loopback reason — docs/platform-flags.md.
+
+  com.apple.security.cs.debugger. Would let this process attach to others. Nothing here debugs
+  anything, and it is the entitlement most worth not having.
+-->
+<plist version="1.0">
+<dict>
+  <!--
+    CoreCLR compiles IL to machine code at runtime and then executes the pages it just wrote. The
+    hardened runtime's default is that no page is both writable and executable, so without this the
+    process does not start — it dies during runtime initialisation, before any of this application's
+    code runs, which means before anything exists that could report it.
+  -->
+  <key>com.apple.security.cs.allow-jit</key>
+  <true/>
+
+  <!--
+    The broader form of the same permission, and it is needed as well as allow-jit rather than
+    instead of it. allow-jit covers pages mapped through the MAP_JIT convention; CoreCLR also
+    allocates executable memory outside that path — stubs, precode, and the write-xor-execute
+    fallback it uses when MAP_JIT is unavailable. With only the first, startup gets further and
+    still fails.
+  -->
+  <key>com.apple.security.cs.allow-unsigned-executable-memory</key>
+  <true/>
+
+  <!--
+    Library validation requires every loaded dylib to be signed by the same team as the main
+    binary. This bundle carries native libraries built by other people — libsodium, libSkiaSharp,
+    libHarfBuzzSharp, libe_sqlite3, libAvaloniaNative — and the release script signs each of them
+    with this Developer ID, which would in principle satisfy validation.
+
+    It is disabled anyway, and the reason is the updater. Velopack replaces the bundle in place and
+    relaunches it, and the process doing the replacing is not always signed by the same team as the
+    process being replaced during the changeover. Leaving validation on makes the failure mode of a
+    bad update "the application will not start", with no way to recover except a reinstall the user
+    would have to be told about through some other channel.
+  -->
+  <key>com.apple.security.cs.disable-library-validation</key>
+  <true/>
+
+  <!--
+    The runtime reads DYLD_ variables while resolving its own native dependencies, and Velopack's
+    update path sets them. Without this the hardened runtime strips them silently and the failure
+    surfaces later as a library that cannot be found, naming a file that is plainly present.
+  -->
+  <key>com.apple.security.cs.allow-dyld-environment-variables</key>
+  <true/>
+</dict>
+</plist>

build/macos/Info.plist.template

@@ -0,0 +1,125 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!--
+  The Info.plist for the macOS bundle, with the version left as a placeholder.
+
+  ◆ A TEMPLATE RATHER THAN A FILE, BECAUSE vpk COPIES A CUSTOM PLIST VERBATIM.
+
+  Measured, not assumed: `vpk [osx] bundle --plist` performs no substitution of any kind. It logs
+  "Bundle using provided Info.plist" and copies the bytes. That is also why it refuses --plist and
+  --bundleId together — with a plist supplied, every key is the caller's problem.
+
+  So a committed Info.plist would carry whatever version it was written with into every release
+  afterwards, and the failure is quiet in the worst way: Velopack's own release index would carry the
+  right version, the updater would compare correctly and update correctly, and only the About window,
+  Finder's Get Info panel and any crash report would claim the build was something else. Nobody
+  reads those on the day of a release. scripts/release-macos.sh substitutes @VERSION@ into a copy
+  and passes that.
+
+  ◆ WHY A CUSTOM PLIST AT ALL, WHEN vpk WRITES A PERFECTLY GOOD ONE.
+
+  Three keys it does not write, each of which is a real defect without it:
+
+    CFBundleDisplayName    The bundle on disk is DodoSSH.Desktop.app, because the pack id must not
+                           be DodoSSH — see scripts/release-macos.sh for the directory collision
+                           that rule prevents. On Windows the pack id is invisible; on macOS it
+                           names the thing in /Applications and in the Dock. This key is what puts
+                           "DodoSSH" back in front of a person while the bundle keeps the id.
+
+    LSMinimumSystemVersion Without it macOS will happily launch this on a release the runtime was
+                           never built for, and the user gets a dyld crash rather than a sentence.
+
+    NSHumanReadableCopyright  Shown in the About panel. Absent, the panel shows a blank line.
+-->
+<plist version="1.0">
+<dict>
+  <!--
+    CFBundleName is what the menu bar shows and is capped at 15 characters by convention;
+    CFBundleDisplayName is what Finder and the Dock show. Both say DodoSSH, and the bundle
+    directory does not. See the note above.
+  -->
+  <key>CFBundleName</key>
+  <string>DodoSSH</string>
+
+  <key>CFBundleDisplayName</key>
+  <string>DodoSSH</string>
+
+  <!--
+    Reverse-DNS under the domain this project actually controls. It is the identity Gatekeeper,
+    the notary service and the keychain all key off, so it is as irreversible as the Windows pack
+    id: changing it makes an update a different application, and it orphans anything the previous
+    identifier stored — including the Secure Enclave key MacDeviceKeyStore holds, which is scoped
+    to this identifier and cannot be migrated because its whole point is that it never leaves the
+    enclave.
+  -->
+  <key>CFBundleIdentifier</key>
+  <string>dev.dodotech.dodossh</string>
+
+  <!--
+    The apphost the publish produced, named for the product by <AssemblyName> in the csproj rather
+    than for the project. Must match --mainExe or the bundle launches nothing.
+  -->
+  <key>CFBundleExecutable</key>
+  <string>DodoSSH</string>
+
+  <!--
+    Both version keys take the numeric core only — 1.2.3 and never 1.2.3-rc.1 — because Apple
+    defines them as one to three dot-separated integers and notarization rejects what it cannot
+    parse. The full version, prerelease suffix and all, is in Velopack's release index, and that
+    is the one the updater compares. These two are for Finder and for Gatekeeper.
+
+    They are the same value rather than the usual marketing/build split, because there is no build
+    counter here that a release does not already bump.
+  -->
+  <key>CFBundleShortVersionString</key>
+  <string>@VERSION@</string>
+
+  <key>CFBundleVersion</key>
+  <string>@VERSION@</string>
+
+  <!--
+    The file name inside Contents/Resources, which is where --icon puts it. With a custom plist
+    nothing rewrites this key, so a rename of the asset that forgets this line produces a bundle
+    showing the generic application icon and no error anywhere.
+  -->
+  <key>CFBundleIconFile</key>
+  <string>dodossh.icns</string>
+
+  <key>CFBundlePackageType</key>
+  <string>APPL</string>
+
+  <!--
+    12.0, and it is read off the binaries rather than off a support matrix. The apphost and
+    libcoreclr.dylib in a net10.0 osx-arm64 publish both carry LC_BUILD_VERSION with minos 12.0.0,
+    so 12.0 is the oldest release these bytes are built to load on.
+
+    Microsoft's *support* statement for .NET 10 is higher than this, and that difference is
+    deliberate rather than overlooked: this key decides whether macOS refuses to launch the app at
+    all, and refusing on a release where it would in fact have run is the worse of the two errors.
+    A user on an unsupported-but-working macOS gets the application; the support matrix governs
+    what gets fixed if it misbehaves there, which is a different question.
+  -->
+  <key>LSMinimumSystemVersion</key>
+  <string>12.0</string>
+
+  <!--
+    Without this the window is drawn at 1x and scaled up, which on a Retina display turns the
+    terminal — the one surface in this application that is nothing but small text — into a blur.
+  -->
+  <key>NSHighResolutionCapable</key>
+  <true/>
+
+  <key>NSPrincipalClass</key>
+  <string>NSApplication</string>
+
+  <!--
+    False, and stated rather than left out. An agent application has no Dock icon and no menu bar;
+    this one is an ordinary windowed application and the default is already false, but the key
+    being absent is indistinguishable from somebody having removed it.
+  -->
+  <key>LSUIElement</key>
+  <false/>
+
+  <key>NSHumanReadableCopyright</key>
+  <string>© DodoTech. MIT licensed.</string>
+</dict>
+</plist>

docs/adr/0007-device-key-protection.md

@@ -1,4 +1,4 @@
-# ADR 0007 — What protects the device key on Windows
+# ADR 0007 — What protects the device key on the desktop
 
 **Status:** accepted, 2026-07-30
 **Supersedes nothing. Constrains** the device-unlock work described in the client roadmap.
@@ -141,6 +141,14 @@ would have become false under DPAPI alone. A gesture is still something the atta
 - **A TPM is not always there.** A machine without one gets a store that reports itself unavailable, so
   unlock keeps asking for the passphrase and neither affordance appears in the interface. The passphrase path
   is therefore required, not a nicety.
+- **macOS reaches the same decision through different hardware, and the argument transfers intact.**
+  `MacDeviceKeyStore` puts the wrapping key in the Secure Enclave under an access control requiring user
+  presence, so Touch ID or the login password is a condition of *using* it and the enforcement is the
+  platform's rather than the process's — which is the entire point of the 2026-07-30 amendment above, and
+  the thing a self-drawn prompt over a protected file would fail to be. The mechanical differences are
+  incidental: P-256 with ECIES because the enclave holds no other kind of key, and no prompt when sealing
+  because the public half needs no consent. See docs/platform-flags.md for the three ordinary Macs where the
+  probe answers no, one of which is every unsigned development build.
 - **The stored key must be treated as losable at any time** — a reset PIN, a cleared TPM, a replaced key.
   Every loss degrades to a passphrase prompt and never to a locked-out vault, which is why every failure in
   the store returns null rather than throwing and why the three unlock statuses all end in the same advice.

docs/adr/0013-desktop-distribution-and-updates.md

@@ -228,6 +228,44 @@ changes.
 token, and it puts a compellable third party in the signing path — which is ADR 0011 rule 3's shape one
 layer down, declined there for reasons that do not stop applying because the vendor changed.
 
+**This rule is Windows-only, and macOS gets the opposite one.** See decision 10: there is no "unsigned for
+now" available on that platform at any price, because Gatekeeper refuses rather than warns.
+
+### 10. macOS is a second desktop platform on the same machinery, signed from the start
+
+The macOS head is the same application, the same Velopack, and the same two-phase person-run release. Four
+things differ, and each is forced rather than chosen.
+
+**Signing is a precondition, not an improvement.** Decision 8's whole argument — one dialog per user per
+lifetime, buy a certificate when a stranger is invited to install — has no macOS equivalent. An
+un-notarized download is refused outright, so the Developer ID certificate and the notarization round trip
+are the price of the package existing. `scripts/release-macos.sh` therefore refuses to run without the
+signing identities, where the Windows script refuses nothing.
+
+**The channels are `osx` and `osx-nightly`, and they are separate for decision 9's reason.** Four channels
+now publish to one repository, and the only thing keeping a Mac from being offered a Windows package is
+that it never reads that index. The macOS nightly channel is named and has no publisher: CI builds and
+bundles the macOS head to prove it still builds, and uploads nothing, exactly as it does for the Windows
+release channel.
+
+**The pack id is shared with Windows, and on macOS it is visible.** vpk names the bundle after the pack id,
+so `/Applications` holds `DodoSSH.Desktop.app`. Decision 2's reasoning applies with more force here rather
+than less: a pack id of `DodoSSH` would put Velopack's install root on `~/Library/Application
+Support/DodoSSH`, which is `ClientPaths.DataDirectory`, and an uninstall would take the user's un-synced
+outbox with it. `CFBundleDisplayName` puts the product name back in front of a person; the directory keeps
+the id.
+
+**arm64 only, because the check is the scarce thing.** Velopack keys a channel to one architecture, and an
+Intel package would be the only artefact in this repository reaching users without somebody having walked
+Phase 18 against it. The engineering for a second channel is small and is described in the release script;
+what is missing is an Intel Mac to verify on, and shipping blind is the thing this project's manual-check
+discipline exists to refuse.
+
+**And one thing that does not differ, which is worth saying because it is the expensive half.** The
+capability to publish still lives on a person's machine and never in CI. Notarization does not change that:
+Apple's ticket says this build came from this developer account, and says nothing about whether the build
+should have been made. Velopack clients still apply what their feed serves. Rule 3 is untouched.
+
 ### 9. There is a second desktop channel, published by CI, and it is a second application
 
 [ADR 0014](0014-android-updates.md) gave the phone a nightly channel and rule 3 above gives the desktop

docs/android-port.md

@@ -30,7 +30,10 @@ verified is that it compiles, links, packages, and carries the right natives.
 transfers protected by a **foreground service**. File transfer is not in the first scope; when it arrives it
 is **one remote pane** with Android's document picker for moving files in and out. *It has since arrived,
 both ways:* the pane, the queue, `ACTION_OPEN_DOCUMENT` going in and `ACTION_CREATE_DOCUMENT` coming out,
-with the foreground service now counting transfers as well as shells.
+with the foreground service now counting transfers as well as shells — and, since, an idle-but-connected
+Files session as well, which a transfer count alone was blind to. *Corrected the same round:* the service's
+other half — a shell's own opening — had never been wired to anything at all, so a shell survived only for
+as long as the app stayed foreground; see [Sessions survive backgrounding](#sessions-survive-backgrounding-via-a-foreground-service).
 
 **What was actually checked**, so the rest can be read with the right amount of trust:
 
@@ -235,6 +238,37 @@ The parts that are definitely different are the on-screen keyboard, and the fact
 needs Ctrl, Esc, Tab and arrows that the software keyboard does not offer — every Android SSH client ships an
 accessory key row for this. That is UI work, not porting.
 
+> **⚠️ Corrected by the build. The data plane assumed a renderer that attaches once and lives forever, and
+> that assumption is WebView2's truth, not Android's.** Desktop's WebView2 process starts with the window and
+> dies with it; `TerminalDataPlane` was written to that reality — one socket, attached once,
+> `Interlocked.Exchange`-guarded against a second attach ever happening at all. On a phone the WebView's own
+> renderer process is a separate thing from the app process the foreground service above is keeping alive,
+> and Android kills *that* independently — under memory pressure, or simply for being backgrounded — with no
+> foreground service able to save it. The page then reloads with a fresh socket, and three things broke on
+> that reload before this was found: the second attach was refused outright (`409 Conflict`), because a
+> second valid upgrade could only mean a bug or a hostile second process, never our own page coming back; a
+> send into the dead first socket threw, and that exception unwound `TerminalSessionPump`'s flush loop,
+> freezing the still-live shell behind it — `LiveSessionCount` kept counting a session nothing would ever
+> drain again; and every byte sent while no page was attached had already spent flow-control credit that no
+> acknowledgement could ever return, so a session outliving 256 KiB of output into a dead page stalled for
+> good regardless of the other two. Waiting for the old socket to notice it was dead and close on its own
+> was never going to be enough either — a killed renderer sends no TCP FIN, so the old receive loop could sit
+> unaware for the whole 30-second keepalive.
+>
+> Fixed as a takeover rather than a guard: a second valid upgrade — origin, token and subprotocol all
+> checked exactly as before — now displaces whatever socket was attached instead of being refused, since
+> only this app's own page ever knows the token, so a second valid attach *is* that page, back again.
+> `TerminalDataPlane.SendAsync` no longer lets a dead-socket send escape as a fault; it reads as "nobody
+> listening," same as no socket being attached at all. `TerminalWorkspace` resets each live session's credit
+> window on every attach and resends its `SessionOpened` frame, flagged as a replay, so the fresh page
+> rebuilds the pane and the pump stops waiting on an acknowledgement that was never coming. And
+> `terminal.js`'s socket now retries itself, forever, with backoff, instead of reporting the connection
+> failed and stopping — the page dies with the app anyway, so there is no case where retrying is the wrong
+> call. What is **not** recovered, and says so rather than pretending otherwise: scrollback across a page
+> reload. It lived in the page's own DOM, and a reloaded page is a new DOM. The replay banner — *"the view
+> reconnected; earlier output stayed on the host"* — is that honesty put where the person looking at the
+> terminal will actually read it, not buried in a log.
+
 ---
 
 ## Decisions taken
@@ -280,13 +314,50 @@ What is desktop-only is the *left* pane — `LocalDirectory`, the drive list, th
 
 ### Sessions survive backgrounding, via a foreground service
 
-A persistent notification for as long as a shell or a transfer is live.
+A persistent notification for as long as a shell, a transfer, or an idle-but-connected Files session is
+live.
 
 It costs the user a notification and some battery. It buys the behaviour the desktop client already promises
 and documents — that a shell outlives a vault lock, and that a transfer finishes — and the alternative was
 to make `TerminalWorkspace`'s guarantee desktop-only, which is a worse thing to have to write down than a
 notification is to look at.
 
+**Three corrections found after the first cut shipped, all in the wiring rather than the design:**
+
+- **A shell opening never started the service.** `SessionKeepAlive` heard `TerminalWorkspace.SessionEnded`
+  and refreshed on that, but nothing announced the opposite event — so a user who opened a shell and
+  backgrounded the app immediately had no foreground service at all, and Android was free to kill the
+  process holding it. `MainWindowViewModel.TerminalSessionOpened` is now wired the same way in
+  `App.axaml.cs`'s `ComposeKeepAlive`.
+- **A connected-but-idle Files session counted as nothing.** A host open on the Files screen with no
+  transfer moving is a live SFTP connection a dying process would sever, and the old two-argument
+  `Reconcile(liveSessions, activeTransfers)` had no way to hear about it. `TransfersViewModel.HasLiveFileSession`
+  — `IsConnected` with a real `ConnectedCipher`, which a bucket never has — is the third fact `Reconcile` now
+  takes.
+- **Refreshing the notification restarted the service, which throws when backgrounded.** `Reconcile` called
+  `StartForegroundService` on every refresh, including the common case of a service that was already
+  running. On API 31+ that throws `ForegroundServiceStartNotAllowedException` the instant the app is
+  backgrounded — a transfer finishing in the pocket, one of two shells dying — which crashed the process and
+  took every session with it. `SessionForegroundService` now tracks whether it is already running and, when
+  it is, posts the updated notification through `NotificationManager.Notify` instead of asking Android to
+  start anything.
+
+**The notification permission is requested, not just declared.** API 33+ requires `POST_NOTIFICATIONS` at
+runtime or the receipt is silently invisible — the service still runs, but nothing on screen says so.
+`SessionForegroundService.Reconcile` asks for it the first time in this process there is actually something
+to show, at most once, with no result read back: a refusal costs the notification and nothing else, which is
+what the manifest's own comment on the permission says.
+
+**And a fourth correction, found by the notification refusing to come down.** "1 shell connected" outlived
+the shell, both ways a shell can close. A shell exiting on its own announced `SessionEnded` from inside its
+run's own finally block — where the run task is by definition not yet complete, so the
+`LiveSessionCount` the keep-alive reads from that event still counted the session that had just ended, and
+nothing fired afterwards to correct it. A tab closed by hand announced nothing at all, by a recorded
+decision that assumed every subscriber was the closer. Both reversed in `TerminalWorkspace`: the end is now
+announced from a continuation after the run has actually completed, and `CloseSessionAsync` announces too,
+after its own drain — the event's remark carries the reversal, and `SessionEnded`'s subscribers were all
+already "reconcile to reality" handlers for which a second announcement is harmless.
+
 ### Phone first
 
 About 360dp wide. The tablet route was cheaper — a landscape tablet is close to the existing 880×560 minimum
@@ -523,7 +594,12 @@ go at 360dp:
    stopping it from a count rather than a lifecycle. `TerminalWorkspace.LiveSessionCount` is the source of
    truth deliberately: it already knows that a session whose shell exited is not live, which a counter
    incremented on open would not, and a phone showing "1 shell connected" over nothing would be exactly the
-   dishonesty the unlock screen's count exists to prevent.
+   dishonesty the unlock screen's count exists to prevent. *Corrected since:* the opened half of a shell's
+   lifecycle was never wired in, so the service could never come up for a shell at all; an idle-but-connected
+   Files session now counts as a third live fact rather than nothing; a refresh while backgrounded updates
+   the notification in place instead of restarting the service, which the API throws on; and
+   `POST_NOTIFICATIONS` is now actually requested rather than merely declared. See
+   [Sessions survive backgrounding](#sessions-survive-backgrounding-via-a-foreground-service) for all four.
 7. ~~**The interface**, phone-first.~~ **Done for the decided scope** — all seven screens of the design,
    plus the two states the design does not draw because it starts at an enrolled phone (naming a server, and
    choosing a passphrase).

docs/design-import-gaps.md

@@ -258,7 +258,7 @@ the chrome, hosts and terminals, file transfer, the vault, teams, and preference
 > | The host card as a link straight to a terminal | Click still selects, double-click still connects, and the pencil still opens the pane — the mock's card-as-link is not adopted, because multi-select (Ctrl, Shift, the marquee band) depends on a plain click meaning "choose this one" rather than "go". |
 > | Quick connect's SSH/SFTP kind column | The auth word — credential, key, or password — see above. |
 > | A collapsed section staying collapsed after a restart | In memory only, for the running session. `settings.json` holds two scalars by decision — the terminal's text size and whether this machine checks for updates on its own — and collapse state is not judged worth a third. |
-> | QUICK ACCESS's editor, on the phone | **Deferred; the data is not.** `HostSecret.PinnedPaths` is shared, synced and merged on both heads, so a pin made on the desktop reaches the phone and back — Android just has nowhere yet to add or remove one itself. |
+> | QUICK ACCESS's editor, on the phone | ◆ **Shipped, over the same shared data the deferral above described.** The phone's host editor draws its own QUICK ACCESS section on the same staged `VaultViewModel.EditorPinnedPaths` the desktop's drawer binds — a row per pin, an add field and button, and a remove target sized to this head's 44dp touch floor rather than the desktop's 22-pixel close box. `AddEditorPinCommand`'s refusals surface through a `Status` line the editor page draws for itself, since that page covers the whole screen and the list behind it draws its own `Status` off-screen for as long as it is open. The pins themselves reach a second surface this head has that the desktop does not need: `TransfersViewModel.ConnectedPinnedPaths` carries them as chips on the Files screen while connected, and tapping one runs `GoRemoteCommand` — the same command the breadcrumb trail already used to navigate. Two deviations from the desktop, both named where they land: the chip row is a snapshot taken at connect rather than a live follow of the vault, so a pin edited mid-session shows up on the next connect rather than this one; and the editor's own hint sentence says the pins appear on the Files screen, not above a terminal — this head has no terminal strip for them to sit above, the same honesty the row below already states for the desktop's own hint. |
 > | Collapse All beside every section's own collapse chevron | Bound on every heading's view model and shown on only the first — `SidebarGroupHeader.IsFirstBoardSection` is what a virtualised list of sections uses in place of a control of the board's own that would otherwise have to sit above all of them. |
 > | The QUICK ACCESS hint's claim that pins live in a sidebar | "Pinned folders appear above the terminal for this host." — no sidebar exists on this screen for the sentence to point at, so the shipped hint says where they actually draw. |
 > | The mock's "Saving to **DodoTech ▾** vault" subtitle, with a picker's chevron inside a sentence | The pre-existing `DrawerSubtitle` wording — the vault's name alone, unchanged by this pass. A chevron inside running text implies the text itself is the control, which it is not: the vault picker is its own element, shown only while creating and only above one writable vault, as it always has been. |

docs/manual-checks.md

@@ -1101,7 +1101,7 @@ time)" — is the worse failure of the two: it rebinds a host as a side effect o
 
 ## Phase 8 — Adding and removing on the phone's host list
 
-Thirteen checks, and the reason there are thirteen rather than none is worth stating: **the layout suite
+Twenty-one checks, and the reason there are twenty-one rather than none is worth stating: **the layout suite
 cannot see any of this and structurally never will.** `DodoSSH.Client.App.Layout.Tests` targets `net10.0` and
 `DodoSSH.Client.Android` targets `net10.0-android`, so a project reference is impossible; Avalonia's
 application, dispatcher and platform are one-shot process globals, so a second head cannot share the
@@ -1418,6 +1418,60 @@ destinations, so switching under a live one would show a screen titled S3 listin
 the row — that screen keeps its own copy of the host list, so it has to be re-found there by entity id
 rather than handed the vault's object.
 
+### 8.18 QUICK ACCESS in the phone's host editor
+
+Open a host's editor and scroll to QUICK ACCESS.
+
+**Pass:** an empty list, an add field and an ADD button. Type `/var/www/app` and press ADD.
+
+**Pass:** a row appears carrying that path and a ✕ at least 44dp on a side. Type the same path again and
+press ADD.
+
+**Pass:** nothing is added, and a sentence appears on the page saying the path is already pinned — this page
+covers the whole screen while it is open, so that sentence is this page's own `Status` line rather than the
+one the host list draws above HOSTS, which is off-screen right now. Clear the box and press ADD with nothing
+typed.
+
+**Pass:** a sentence saying a pinned path cannot be blank, in the same place.
+
+Press the ✕ on the pinned row, then SAVE.
+
+**Pass:** back on HOSTS with the pin gone. Open the editor on that host again.
+
+**Pass:** QUICK ACCESS is empty — the removal was saved, not merely staged. Re-pin `/var/www/app`, SAVE, and
+check the same host on the desktop.
+
+**Pass:** the pin is there. `HostSecret.PinnedPaths` is shared and merged like every other field on a host,
+so nothing about this page keeps its own copy.
+
+**Failure means:** a refusal that changes nothing on screen is `AddEditorPinCommand` writing to `Status`
+with nothing on this page bound to it — the honesty rule broken silently, since the command still behaves
+correctly and only the telling of it is missing. A pin gone after SAVE-then-reopen but present on the
+desktop is `BuildHost` not reading `EditorPinnedPaths`, or `EditSelectedHost` not loading it back in.
+
+### 8.19 Pin chips on the Files screen
+
+Pin a folder on a host, then connect to it on the files screen (SFTP), either directly or via **Connect via
+SFTP**.
+
+**Pass:** once connected, a row of chips appears between the breadcrumb and the listing, one per pin, each
+at least 44dp tall. Tap one.
+
+**Pass:** the listing navigates straight to that directory, the same as tapping a breadcrumb crumb does.
+Disconnect, then connect to a host with nothing pinned.
+
+**Pass:** no chip row at all — not an empty one. Connect to a bucket instead.
+
+**Pass:** still no chip row, on any bucket. A bucket has no `HostSecret` underneath it and so nothing to
+pin.
+
+**Failure means:** chips that do not move the listing are the row's `GoRemoteCommand` binding pointed at the
+wrong `DataContext` — see the `$parent[views:FilesScreen]` escape every other command in this file uses. A
+chip row surviving a disconnect, or appearing under a bucket, is `TransfersViewModel.ConnectedPinnedPaths`
+not being cleared in `CloseSessionAsync` or `OpenBucketAsync`. A chip row missing a pin added *after* this
+connect is not a bug — see `ConnectedPinnedPaths`'s own remark on why this is a snapshot rather than a live
+follow, and try disconnecting and reconnecting instead.
+
 ---
 
 ## Phase 9 — Tag chips and the picker
@@ -1691,6 +1745,57 @@ is a terminal that answers the buttons and ignores the keyboard: it reads as the
 Worth doing on the software keyboard too, where the same fault shows as the keyboard closing on the first
 tap of an arrow key.
 
+### 11.10a The accessory keys do not cost the terminal its *software* keyboard either
+
+With a shell open and the software keyboard up, tap **esc**, **tab** or an arrow on the accessory row, then
+keep typing on the software keyboard.
+
+**Pass:** the keyboard settles back unchanged — same layout, same suggestion strip, same height — and
+everything typed after the tap still reaches the terminal. The accessory row stays visible above the
+keyboard throughout. A blink during the press itself is tolerable: the platform takes the focus on both
+halves of every touch and the return is posted right behind each theft, so the connection can visibly flap
+for the press's own duration — what it must never do is *stay* swapped after the finger lifts.
+
+**Failure means:** Android's own view focus stayed on Avalonia's input view after the tap instead of being
+handed back. This is the half `Focusable = false` cannot reach — the platform requests focus for its own
+view after dispatching every handled touch — and the symptom chain is the keyboard swapping to its no-input
+layout and the inset churn parking it over the very row that was tapped. The first fix for this failed by
+timing alone: it handed focus back from inside the very dispatch the platform re-steals it after. See
+`TerminalFocus` in the Android head's Platform folder for both the mechanism and the fix's shape.
+
+### 11.11 Closing a connection and opening a new one both take you somewhere real
+
+Open a shell, close its tab, then open a different one from HOSTS.
+
+**Pass:** the new terminal renders and takes input straight away — no stuck "Connecting…" status, no blank
+pane that never receives the prompt.
+
+**Failure means:** `TerminalDataPlane` refused the page's reattach. The renderer's `WebSocket` does not
+survive a tab going from one to zero and back to one on every device, and a host that answers a second valid
+upgrade with `409 Conflict` instead of taking the socket over leaves every terminal after the first
+permanently unreachable — see the correction in `docs/android-port.md`'s terminal section.
+
+### 11.12 A backgrounded shell survives its renderer being killed · **needs several minutes, or developer tooling**
+
+With a shell open and something worth reading in its scrollback, background the app (home button, not back)
+for several minutes — long enough for Android to consider reclaiming it — then return. If the device exposes
+it, forcing a stop of the WebView renderer process from Developer Options while backgrounded is the more
+reliable way to trigger the same thing on demand rather than waiting on the OS's own judgement. Either way,
+type something once you are back.
+
+**Pass:** one of two honest outcomes, both good. Either the pane is exactly as it was — the renderer process
+survived, so nothing needed to happen — or the pane is empty but for a dim line reading `── the view
+reconnected; earlier output stayed on the host ──`, meaning the page reloaded and reattached. In both cases
+what is typed now reaches the shell, and the shell is still the same one — not a new tab, not a reconnect
+sheet, no "Connecting…" status stuck on screen.
+
+**Failure means:** if the status stays stuck or nothing typed arrives, the renderer's socket did not retry
+itself — see `terminal.js`'s `connect()` and its backoff. If the pane came back empty with **no** banner, a
+session that survived a reload is being shown as though its scrollback had too, which is not true and is
+worse than saying nothing: the banner exists so this is never silently wrong. If typing does nothing but the
+banner is there, the session's credit window was not reset on reattach and the shell is frozen behind it —
+see `TerminalWorkspace.ReplayAfterAttachAsync`.
+
 ---
 
 ## Phase 12 — Shared vaults: the operations that span two accounts
@@ -2021,9 +2126,63 @@ Queue several files in each direction, put the phone to sleep with the screen of
 notification goes away when the last one does — with no shell open. With a shell open it stays, because that
 is what it was already for.
 
+Now, separately: open a shell to the host, press the home button (backgrounding rather than sleeping — the
+distinction matters, because backgrounded is the state in which Android is free to kill a process no
+foreground service is protecting), wait thirty seconds with the shell doing nothing, and return.
+
+**Pass:** the notification stayed up the whole time, and the shell is exactly where it was — same scrollback,
+same prompt — with typing reaching the host immediately. Exit the shell.
+
+**Pass:** the notification goes with it, once nothing else is open. Open another shell and close it from the
+shells strip's ✕ instead of exiting — the notification comes down for that route too, which is the route
+that used to leave it up: a deliberate close announced nothing to the keep-alive at all, and a shell exiting
+on its own was announced while the count still included it.
+
 **Failure means:** an upload that stalls with the screen off is the count not reaching
 `SessionForegroundService`, and Android has stopped the process mid-transfer. A notification left up
-afterwards is `ActivityChanged` not being subscribed — the other end of the same wire.
+afterwards is `ActivityChanged` not being subscribed — the other end of the same wire. A shell that has
+disconnected on return is `MainWindowViewModel.TerminalSessionOpened` never reaching `SessionKeepAlive` — the
+service only ever heard about a shell *ending*, so it never came up for one in the first place. A
+notification still saying "1 shell connected" after the shell is gone — by either route — is
+`TerminalWorkspace.SessionEnded` firing before the run completed, or a close not announcing; see
+`AnnounceEndedAsync` and the event's own remark.
+
+### 14.6a A Files connection with nothing moving still survives backgrounding
+
+Connect to a host on the Files screen with no transfer queued — just browse to somewhere and stop. Note the
+directory shown, then background the app, wait thirty seconds, and return.
+
+**Pass:** the notification stayed up the whole time (check the shade if the return is too quick to see it
+directly), and the pane is exactly where it was — the same listing, the same breadcrumb — with no reconnect
+needed.
+
+**Failure means:** `TransfersViewModel.HasLiveFileSession` not reaching `SessionKeepAlive`, so an idle but
+still-open SFTP connection read as nothing running at all and the process was free to die under it.
+
+### 14.6b The notification permission is asked for once, at the first thing worth showing · **needs Android 13+**
+
+On a device running Android 13 or later, on a fresh install that has never connected to anything, open a
+shell or the Files screen for the first time.
+
+**Pass:** a system dialogue asking to allow notifications appears at that moment — not at launch, and not
+before this first connect. Answer it either way; the connection completes regardless, and background the app
+afterwards to confirm nothing else changed about it.
+
+**Failure means:** the dialogue appearing at launch is asking before there is anything on screen to justify
+it. Never appearing at all on API 33+ is the harder failure to notice, because nothing else surfaces it —
+the service still starts and still holds the process open, only the receipt is invisible. See
+`SessionForegroundService.RequestNotificationPermission`.
+
+### 14.6c Refusing the permission costs the notification and nothing else
+
+Continuing from 14.6b: choose **Don't allow** on the system dialogue. Queue a transfer, or open a shell, and
+background the app.
+
+**Pass:** no notification appears anywhere, but the transfer still finishes, or the shell is still there on
+return, exactly as in 14.1–14.6a.
+
+**Failure means:** anything disconnecting or failing here is the permission refusal being read as though it
+had refused the service itself, rather than only the notification Android draws for it.
 
 ### 14.7 SAVE FILE writes where you pointed it, and the file opens
 
@@ -2437,3 +2596,120 @@ package manager will not offer to.
 
 **Failure means:** the channels are not separate, and a public key is signing the application people keep
 their credentials in.
+
+## Phase 18 — Installing the macOS client, and being updated by it
+
+The macOS counterpart of phase 16, and it needs a Mac with a Secure Enclave — an Apple Silicon machine or
+an Intel one with a T2. Every check here is structurally unreachable by a test for the reasons phase 16
+gives, plus one this platform adds: **CI has no macOS runner at all**, so this phase is the only place the
+suite and the application ever run on macOS. Anything `docs/platform-flags.md` marks as unverified on macOS
+is verified here or nowhere.
+
+Run `bash scripts/release-macos.sh` first. It stops after packing and notarizing, on purpose, so that
+everything below happens before anything reaches a user. Phase 16.0 — the feed being readable without
+credentials — applies unchanged and is not repeated.
+
+### 18.1 Gatekeeper accepts it on a machine that did not build it · **do this one first**
+
+The Mac that signed a package trusts it locally whatever happened, so the build machine cannot answer this
+question about itself. Copy the `.pkg` to a second Mac — or at minimum download it through a browser, which
+is what applies the quarantine attribute — and open it.
+
+**Pass:** it installs with no warning beyond the ordinary installer prompts.
+
+**Failure means:** "cannot be opened because Apple cannot check it for malicious software" is notarization
+that did not happen or a ticket that did not staple. The script's `spctl --assess` and `xcrun stapler
+validate` should have caught it before this point, so reaching here means one of those two checks was
+removed or skipped. Do not distribute the package.
+
+### 18.2 The Dock shows the product and not the pack id
+
+Look at the installed application in `/Applications`, in the Dock, and in the menu bar while it runs.
+
+**Pass:** the menu bar says **DodoSSH**. Finder shows **DodoSSH**. The bundle on disk is
+`DodoSSH.Desktop.app` and that is expected — see the pack id note in `scripts/release-macos.sh`.
+
+**Failure means:** "DodoSSH.Desktop" in the menu bar is `CFBundleName` not reaching the bundle, which means
+the rendered `Info.plist` did not get used. Since vpk copies a custom plist verbatim and substitutes
+nothing, check the same bundle's `CFBundleShortVersionString` — if it reads `@VERSION@`, the template was
+passed through unrendered.
+
+### 18.3 The icon is the mark, at every size
+
+Look at it in the Dock, in Finder's icon view at a large size, and in `⌘I` Get Info.
+
+**Pass:** the accent tile and the `>_` mark, crisp at 1024, with the same air around it that Finder and
+Safari have.
+
+**Failure means:** a generic application icon is `CFBundleIconFile` naming a file that is not in
+`Contents/Resources`. An icon that fills its square edge to edge, larger than its neighbours, is
+`New-MarkPng` having been called with the Windows tile fraction — see `dodossh-icon.ps1`.
+
+### 18.4 Touch ID guards the device key, and the enclave enforces it
+
+Register a device key from the security settings page, then lock the vault and unlock it again.
+
+**Pass:** registering shows **no** prompt at all — sealing uses only the public half — and unlocking raises
+the system Touch ID sheet saying DodoSSH is trying to *unlock your DodoSSH vault*. The vault opens on a
+successful touch.
+
+**Failure means:** a prompt at registration is not a failure of correctness but says the key was not created
+in the enclave; check that `kSecAttrTokenID` reached the attributes. **No prompt at unlock, with the vault
+opening anyway, is the serious one** — it means the key is a software key and the access control did nothing,
+which is precisely the "a gate inside the process is not a gate" mistake `WindowsDeviceKeyStore` documents.
+
+### 18.5 Declining the fingerprint falls back to the passphrase
+
+Repeat 18.4 and cancel the Touch ID sheet.
+
+**Pass:** the unlock screen asks for the passphrase, and it works.
+
+**Failure means:** an error dialog, or a stuck screen, is `TryLoadAsync` throwing rather than answering
+null. Every failure it can meet — cancelled, timed out, key invalidated by a password reset — is meant to
+be indistinguishable and to land on the passphrase.
+
+### 18.6 A development build offers no device key at all
+
+Run the application with `dotnet run` rather than from the installed bundle, and open the security settings
+page.
+
+**Pass:** registering a device key is not offered.
+
+**Failure means:** being offered it is `IsSupported` having inferred availability from the OS rather than
+probing. An unsigned build cannot create an enclave key, so accepting the offer would put a wrap on the
+server that nothing can ever open and list a capability this machine does not have.
+
+### 18.7 The terminal works, which is the WKWebView question
+
+Connect to a host and use the shell: type, run something that scrolls, resize the window.
+
+**Pass:** the terminal attaches within a second or two and behaves as it does on Windows.
+
+**Failure means:** a blank pane that reports a renderer timeout after fifteen seconds is the loopback
+WebSocket not reaching WKWebView. This is the check that most needs walking, because the data plane has
+never run against this backend — see `TerminalDataPlane`. If it fails, the App Sandbox is the first thing to
+rule out: the entitlements deliberately do not enable it, and a sandboxed process cannot listen on loopback
+without `com.apple.security.network.server`.
+
+### 18.8 An update is offered, downloaded and applied
+
+With the release installed, cut a second release with a higher version and publish it, then leave the first
+running.
+
+**Pass:** the banner appears, downloads, and on applying the application closes and reopens on the new
+version. The vault's contents and the known hosts survive.
+
+**Failure means:** an update that never arrives is usually the channel — `osx` here and `osx` in
+`VelopackUpdateChannel.MacReleaseChannel`, with no error anywhere when they disagree. An update that
+downloads and fails to apply, leaving the application unable to restart, is library validation: check that
+`com.apple.security.cs.disable-library-validation` survived into the entitlements.
+
+### 18.9 Uninstalling does not take the vault with it
+
+Register a device, sync something, then remove the application.
+
+**Pass:** `~/Library/Application Support/DodoSSH` still holds the cache and the outbox afterwards.
+
+**Failure means:** an empty directory is the pack id having been changed to `DodoSSH`, which puts Velopack's
+install root on top of `ClientPaths.DataDirectory` and makes an uninstall delete a user's un-synced work.
+This is the single reason the bundle is named `DodoSSH.Desktop.app`.

docs/platform-flags.md

@@ -3,8 +3,18 @@
 Things known or suspected to behave differently outside Windows, plus deployment gotchas that
 have already cost time once. Development is Windows-first, but **the full test suite now runs on
 Linux in CI on every change**, so a Linux claim here is usually a measurement now rather than a
-suspicion. **macOS is still untested**, and anything marked *unverified* has not run on the platform
-in question and must not be assumed to work.
+suspicion. Anything marked *unverified* has not run on the platform in question and must not be
+assumed to work.
+
+**macOS now builds and packages, and has still never run.** The distinction matters more here than
+anywhere else on this page, because the two halves are verified in completely different places. The
+build is measured on every main and tag build: CI publishes `osx-arm64` and runs `vpk [osx] bundle`
+on a Linux runner, which is enough to catch a restore graph with no macOS native asset and an `.app`
+that will not compose. Everything past that — whether the window draws, whether the terminal's
+loopback WebSocket reaches WKWebView, whether the Secure Enclave holds a device key — is verified
+only by a person walking Phase 18 of [manual-checks.md](manual-checks.md) on a Mac, because **there
+is no macOS runner in CI**. Treat every macOS runtime claim below as unverified unless it says
+otherwise.
 
 Each entry says what the risk is, why it matters, and what to do about it. Delete an entry when it
 has been verified or made moot — not when it merely stops being convenient.
@@ -17,6 +27,19 @@ docs/crypto.md §1. *Already mitigated* — but if a BCL AEAD path is ever added
 **must** gate on `IsSupported` rather than assuming availability, or the client will fail to open
 any vault on macOS.
 
+**The Secure Enclave holds P-256 keys and nothing else**, which is why `MacDeviceKeyStore` wraps the
+device key with ECIES rather than with the RSA-OAEP the Windows store uses. It will not hold an RSA
+key at any size, so this is not a preference. The useful consequence is that the macOS shape is
+*better* than the Windows one: `SecKeyCopyPublicKey` works on an enclave key without prompting, so
+registering a device is silent and only unlock asks — where Windows raises a dialog at key creation
+too. *Unverified:* no enclave call in this repository has ever run.
+
+**Three ordinary Macs have no usable enclave**, and `IsSupported` probes rather than infers for that
+reason: an Intel machine without a T2, a machine with no login password set, and — the one that
+surprises people — **any build that is not code signed**, because enclave key creation needs a
+signing identity. So `dotnet run` correctly offers no device key at all. Do not "fix" this by
+checking the OS instead; the offer would then put a wrap on the server that nothing can ever open.
+
 **Argon2id timings are measured on one Windows machine only.** 256 MiB with t=4 took 323 ms here.
 The floor and ceiling in `EnrollmentLimits` were chosen against that number. *Unverified
 elsewhere:* recalibrate on the slowest target platform before recommending a default profile,
@@ -26,7 +49,11 @@ and the parameters are stored per user at enrollment, so a bad default is a per-
 **libsodium ships native binaries per RID.** This complicates single-file and AOT publishing, and
 on macOS every native library (`libsodium`, `libSkiaSharp`, `libHarfBuzzSharp`, `libe_sqlite3`)
 must be signed **individually** with `--options runtime --timestamp` before the bundle is signed,
-or notarization fails with an error that does not name the offending file.
+or notarization fails with an error that does not name the offending file. *Mitigated* in
+`scripts/release-macos.sh`, which signs every `.dylib` and `createdump` in a loop before vpk touches
+anything — vpk's own pass uses `codesign --deep`, which is the shape Apple documents as wrong for
+nested code and is the likeliest source of that unnamed rejection. The loop looks redundant next to
+`--deep` and is not; do not delete it because a release once succeeded without it.
 
 ## Desktop client
 
@@ -366,6 +393,37 @@ AppContainer where loopback connections are blocked without a `CheckNetIsolation
 terminal data plane *is* a loopback WebSocket, so MSIX would break the product outright. Velopack
 for Windows/macOS/AppImage; Flatpak and deb/rpm defer updates to the package manager.
 
+**The App Sandbox is ruled out on macOS for the same reason, and the entitlements say so.** A
+sandboxed process cannot listen on loopback without `com.apple.security.network.server`, and the
+terminal is that listener. Developer ID distribution outside the App Store does not require the
+sandbox, so this costs nothing today — but it does mean the Mac App Store is closed to this
+application without solving the data plane differently first. See
+`build/macos/DodoSSH.entitlements`.
+
+**The hardened runtime is not optional and .NET needs four holes punched in it.** Notarization
+refuses a Developer ID submission without it, and CoreCLR will not start under it without
+`allow-jit` and `allow-unsigned-executable-memory` — both, not either, because the runtime allocates
+executable memory outside the `MAP_JIT` path as well. `disable-library-validation` and
+`allow-dyld-environment-variables` are needed for Velopack's updater rather than for the runtime.
+Each is argued individually in the entitlements file; the failure mode for a missing one is a
+process that dies during runtime initialisation, before anything exists that could report it.
+
+**`vpk` cross-compiles to macOS only as far as the bundle.** `vpk [osx] bundle` runs anywhere and
+produces a real `.app`; there is no `[osx] pack` off a Mac, because pack drives `codesign`,
+`notarytool` and `stapler`. So CI can prove the bundle builds and only a Mac can produce something
+installable. Note this is the *opposite* of the Windows story, where `vpk [win] pack` builds the
+whole installer on Linux — the asymmetry is Apple tooling, not a Velopack limitation.
+
+**A custom `Info.plist` is copied verbatim by vpk, with no substitution whatsoever.** That is why
+`--plist` and `--bundleId` are mutually exclusive, and why `build/macos/Info.plist.template` is a
+template the release script renders rather than a committed file. A committed plist would carry one
+version into every release afterwards, and the symptom is silent: Velopack's index would still be
+right, the updater would still work, and only Get Info and any crash report would disagree.
+
+**macOS app icons live on an 824-in-1024 grid.** An icon that bleeds to the edge of its canvas is
+not bolder, it is the one icon in the Dock that is too big. `dodossh-icon.ps1` draws the `.icns` at
+that fraction and the `.ico` at full bleed, from one geometry.
+
 *Checked rather than assumed, now that Velopack is actually wired up:* its Windows path does not
 reintroduce the thing MSIX was ruled out for. `Setup.exe` is an ordinary Win32 executable that unpacks a
 directory under `%LOCALAPPDATA%` and creates shortcuts — there is no `AppxManifest`, no package identity,

scripts/release-macos.sh

@@ -0,0 +1,404 @@
+#!/usr/bin/env bash
+#
+# Builds, packages and publishes the macOS desktop client.
+#
+# The counterpart of scripts/release-windows.ps1, and deliberately the same shape: run by a person, on a
+# Mac that is not a CI runner, in two phases with the upload withheld until somebody has installed what
+# phase one built and walked the manual checks. docs/adr/0011-android-distribution.md rule 1 puts the
+# capability to ship somebody a build on a machine which is not a runner, and
+# docs/adr/0013-desktop-distribution-and-updates.md explains why the token that writes a Gitea release is
+# that capability: Velopack clients trust their feed and do not verify a package signature when they apply
+# it, so whoever can write a release can ship an update every install runs.
+#
+#   1. Without --upload: builds, signs, notarizes, packs, and stops. Nothing has left this machine
+#      except the notarization submission, which Apple sees and users do not.
+#   2. With --upload: asks for the forge token and publishes what phase one produced. It does not
+#      rebuild, so the bytes that reach users are the bytes that were installed and checked.
+#
+# ◆ WHAT IS DIFFERENT FROM THE WINDOWS SCRIPT, AND WHY.
+#
+# Signing is not optional here. On Windows an unsigned installer costs a SmartScreen dialog once per
+# user, which is why that script has no --signParams and says so. On macOS an un-notarized download is
+# refused outright by Gatekeeper — not warned about, refused — so the Developer ID certificate and the
+# notarization round trip are the price of the package being installable at all, not an improvement to
+# be bought later.
+#
+# ◆ CREDENTIALS COME FROM THE KEYCHAIN AND THE ENVIRONMENT, NOT FROM THIS FILE.
+#
+# Three values are read from the environment, and none of them is itself a secret — they name things the
+# keychain holds, and the keychain is what guards the private key and the App Store Connect credentials:
+#
+#   DODOSSH_SIGN_APP_IDENTITY      e.g. "Developer ID Application: DodoTech (TEAMID)"
+#   DODOSSH_SIGN_INSTALL_IDENTITY  e.g. "Developer ID Installer: DodoTech (TEAMID)"
+#   DODOSSH_NOTARY_PROFILE         the profile name given to `xcrun notarytool store-credentials`
+#
+# `security find-identity -v -p codesigning` lists the first two exactly as codesign wants them. The
+# third is created once per machine:
+#
+#   xcrun notarytool store-credentials DodoSSH \
+#     --apple-id you@example.com --team-id TEAMID --password <app-specific-password>
+#
+# The forge token is the one real secret, and it is prompted for rather than read from a file or the
+# environment, and only in the phase that needs it — for the reason the Windows script gives: the fewer
+# minutes a credential that can publish an update spends in a shell's memory the better.
+#
+# Usage:
+#   bash scripts/release-macos.sh
+#   bash scripts/release-macos.sh --upload
+#   bash scripts/release-macos.sh --skip-tests
+
+set -euo pipefail
+
+UPLOAD=0
+SKIP_TESTS=0
+
+for arg in "$@"; do
+    case "$arg" in
+        --upload)     UPLOAD=1 ;;
+        --skip-tests) SKIP_TESTS=1 ;;
+        *)
+            echo "Unknown argument: $arg" >&2
+            echo "Usage: bash scripts/release-macos.sh [--upload] [--skip-tests]" >&2
+            exit 1
+            ;;
+    esac
+done
+
+# ---- The contract with every installed client ---------------------------------------------------------
+
+# Velopack's identity for this application, and it is effectively irreversible for the reasons the Windows
+# script states — it is what an installed client matches an update against.
+#
+# ◆ THE SAME PACK ID AS WINDOWS, AND ON THIS PLATFORM IT IS VISIBLE.
+#
+# vpk names the bundle after the pack id, so this produces DodoSSH.Desktop.app rather than DodoSSH.app,
+# and that is what somebody sees in /Applications. It is kept anyway, because the alternative is worse:
+# a pack id of DodoSSH would put Velopack's install and its uninstall on ~/Library/Application Support/
+# DodoSSH, which is exactly where ClientPaths keeps the encrypted cache, the outbox of changes not yet
+# pushed and the device key. Sharing that directory would mean an uninstall silently taking a user's
+# un-synced work with it. The same reasoning, and the same conclusion, as the Windows script.
+#
+# What a person actually reads is CFBundleDisplayName, which build/macos/Info.plist.template sets to
+# DodoSSH. So the bundle keeps the id and the Dock shows the product.
+PACK_ID='DodoSSH.Desktop'
+PACK_TITLE='DodoSSH'
+PACK_AUTHORS='DodoTech'
+
+# The project's own forge. Never a DodoSSH deployment — ADR 0011 rule 2. The same URL is a constant in
+# VelopackUpdateChannel, and the two have to agree or the client polls somewhere nothing is published.
+# The owner is part of it: Gitea left a 301 at the old organisation's path, which a GET follows and an
+# upload does not.
+REPO_URL='https://git.dodotech.cloud/DodoTech-Public/DodoSSH'
+
+# A contract with VelopackUpdateChannel.MacReleaseChannel. Velopack's macOS default is also "osx", so
+# leaving it unsaid on both sides would work — but unsaid here and stated there is how a feed goes quiet
+# with no error at all: the client checks, finds nothing, and reports itself up to date forever.
+CHANNEL='osx'
+
+# ◆ ARM64 ONLY, AND THAT IS A DECISION RATHER THAN AN OVERSIGHT.
+#
+# Velopack keys a channel to one architecture, so shipping Intel too means a second channel, a second
+# publish, a second set of deltas and a second thing to keep in step with the client's channel picker.
+# That is all affordable. What is not currently affordable is testing it: nobody here has an Intel Mac,
+# and docs/manual-checks.md exists because this project does not ship desktop builds no one has run.
+# An x64 package built blind and published beside a checked arm64 one would be the only artefact in this
+# repository that reached users unverified.
+#
+# Adding it later is this constant, a second channel name in VelopackUpdateChannel, and a picker keyed on
+# RuntimeInformation.ProcessArchitecture — which reports X64 for a build running under Rosetta, so an
+# Intel build correctly stays on the Intel feed. The work is small; the check is the part that is missing.
+RUNTIME='osx-arm64'
+
+REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
+PROJECT="$REPO_ROOT/src/DodoSSH.Client.App/DodoSSH.Client.App.csproj"
+SOLUTION="$REPO_ROOT/DodoSSH.slnx"
+PUBLISH_DIR="$REPO_ROOT/publish/$RUNTIME"
+RELEASES_DIR="$REPO_ROOT/Releases"
+ICON="$REPO_ROOT/src/DodoSSH.Client.App/Assets/dodossh.icns"
+ENTITLEMENTS="$REPO_ROOT/build/macos/DodoSSH.entitlements"
+PLIST_TEMPLATE="$REPO_ROOT/build/macos/Info.plist.template"
+
+write_step() { printf '\n\033[36m==> %s\033[0m\n' "$1"; }
+stop_with()  { printf '\n\033[31m%s\033[0m\n' "$1" >&2; exit 1; }
+
+# ---- Is this machine able to do the job at all? -------------------------------------------------------
+
+if [ "$(uname -s)" != 'Darwin' ]; then
+    # codesign, notarytool and stapler are Apple tooling and exist nowhere else. The build and even the
+    # .app bundle cross-compile fine from Windows or Linux — `vpk [osx] bundle` does exactly that, and
+    # ci.yml uses it to prove the bundle still builds — but a signed, notarized, installable package
+    # cannot be produced anywhere but here.
+    stop_with 'This builds a signed macOS package and has to run on macOS.'
+fi
+
+for tool in dotnet git xcrun codesign; do
+    command -v "$tool" >/dev/null 2>&1 || stop_with "$tool is not on PATH."
+done
+
+# Checked before anything is built rather than at the step that uses them. Notarization is the last thing
+# this script does and the slowest, and discovering there that a profile name was never exported means
+# throwing away a full build and test run.
+for required in DODOSSH_SIGN_APP_IDENTITY DODOSSH_SIGN_INSTALL_IDENTITY DODOSSH_NOTARY_PROFILE; do
+    if [ -z "${!required-}" ]; then
+        stop_with "$required is not set. See the header of this script for what the three are and how to make them."
+    fi
+done
+
+cd "$REPO_ROOT"
+
+# ---- What is being released ---------------------------------------------------------------------------
+
+# Restored before the version is read, and both halves are load-bearing — the same two traps the Windows
+# script documents. -t:MinVer, because -getProperty alone evaluates the project and runs no targets, while
+# MinVer sets Version from inside one, so the read would answer the SDK's default 1.0.0 regardless of the
+# tag. And a restore first, because naming a target that arrives with a package fails MSB4057 on a clean
+# clone where obj/ has no MinVer targets to import yet.
+write_step 'Restoring the desktop head, so the version can be read'
+dotnet restore "$PROJECT" --locked-mode || stop_with 'Restore failed.'
+
+VERSION="$(dotnet msbuild "$PROJECT" -getProperty:Version -t:MinVer -nologo | tr -d '[:space:]')"
+[ -n "$VERSION" ] || stop_with 'Could not read the version from MSBuild.'
+
+TAG="v$VERSION"
+
+# Apple's two version keys take one to three dot-separated integers and nothing else, so a prerelease
+# version has to have its suffix removed before it reaches the plist. 1.2.3-rc.1 becomes 1.2.3.
+#
+# The full version, suffix and all, is what vpk packs and what the release index carries, so the updater
+# still tells an rc from the release it precedes. These two keys are for Finder and Gatekeeper, which
+# care that the string parses and not what it says. See build/macos/Info.plist.template.
+PLIST_VERSION="${VERSION%%-*}"
+PLIST_VERSION="${PLIST_VERSION%%+*}"
+
+write_step "DodoSSH $VERSION  ($PACK_ID, channel $CHANNEL, $RUNTIME)"
+
+# ---- Phase 2: publish what phase 1 built --------------------------------------------------------------
+
+if [ "$UPLOAD" -eq 1 ]; then
+    # The installer package is the artefact a person downloads, so its absence is the honest test of
+    # whether phase one ever ran. A directory holding only a .nupkg is a pack that failed part way.
+    if ! ls "$RELEASES_DIR"/*.pkg >/dev/null 2>&1; then
+        stop_with "Nothing to upload: $RELEASES_DIR has no .pkg. Run this without --upload first."
+    fi
+
+    echo "About to publish the contents of $RELEASES_DIR to $REPO_URL as $TAG."
+    echo 'Only do this once you have installed it and walked Phase 18 of docs/manual-checks.md.'
+
+    # -s so the token is never echoed and never lands in the shell's history.
+    printf 'Gitea token (write:repository): '
+    read -r -s TOKEN
+    echo
+
+    [ -n "$TOKEN" ] || stop_with 'No token given.'
+
+    # --merge because Gitea already has a release entry for the pushed tag — and on this platform it may
+    # also already hold the Windows package for the same tag, which is the case --merge is really doing
+    # the work for: without it the second platform to publish a given version fails on a release that
+    # exists, and with it the two sit side by side under one tag. --channel keeps the indexes apart.
+    UPLOAD_ARGS=(
+        upload gitea
+        --repoUrl "$REPO_URL"
+        --token "$TOKEN"
+        --outputDir "$RELEASES_DIR"
+        --channel "$CHANNEL"
+        --releaseName "$TAG"
+        --tag "$TAG"
+        --merge
+        --publish
+    )
+
+    # Mirrors the rule the docker image job and the Windows script already apply to the same tag, so a
+    # release candidate is a prerelease in every channel or in none.
+    case "$VERSION" in
+        *-*) UPLOAD_ARGS+=(--pre) ;;
+    esac
+
+    write_step 'Uploading'
+    dotnet vpk "${UPLOAD_ARGS[@]}" || stop_with 'vpk upload failed.'
+
+    write_step "Published $TAG."
+    exit 0
+fi
+
+# ---- Phase 1: build, sign, notarize, pack -------------------------------------------------------------
+
+[ -z "$(git status --porcelain)" ] || stop_with 'The working tree is not clean. A release is cut from a commit, not from a desk.'
+
+HEAD_TAG="$(git describe --exact-match --tags HEAD 2>/dev/null || true)"
+[ -n "$HEAD_TAG" ] || stop_with "HEAD is not tagged. Tag it $TAG first, or change the version and tag that."
+
+# Cannot happen while MinVer is deriving the version from this very tag, and checked anyway: the day
+# somebody pins a version by hand this is the guard that notices.
+[ "$HEAD_TAG" = "$TAG" ] || stop_with "HEAD is tagged $HEAD_TAG but the computed version is $VERSION."
+
+write_step 'Restoring tools'
+dotnet tool restore || stop_with 'dotnet tool restore failed.'
+
+write_step 'Restoring packages (locked, exactly as CI does)'
+dotnet restore "$SOLUTION" --locked-mode || stop_with 'Restore failed. A lock file that only works on Linux fails here.'
+
+write_step 'Building'
+dotnet build "$SOLUTION" --no-restore --configuration Release || stop_with 'Build failed.'
+
+if [ "$SKIP_TESTS" -eq 0 ]; then
+    # The end-to-end suite starts containers and takes minutes. It is run here anyway rather than taken
+    # on trust from CI, because a tag is the one build nobody is watching — and on this platform there is
+    # a second reason: CI has no macOS runner, so this is the only place the suite ever runs on a Mac at
+    # all. Everything docs/platform-flags.md lists as unverified on macOS is verified here or nowhere.
+    write_step 'Testing'
+    dotnet test "$SOLUTION" --no-build --configuration Release || stop_with 'Tests failed.'
+fi
+
+write_step "Publishing $RUNTIME"
+rm -rf "$PUBLISH_DIR"
+
+# Self-contained, and not single-file, for the reasons the Windows script gives: the native libraries ship
+# per RID and a self-extracting bundle breaks delta updates.
+#
+# RestoreLockedMode=false, and the lock files put back straight afterwards. A RID-specific publish resolves
+# a graph the committed lock files do not describe, because they are deliberately kept RID-free —
+# declaring a RID on the head writes a net10.0/<rid> target into every project it references transitively,
+# including DodoSSH.Contracts and DodoSSH.Crypto, and the API's Dockerfile then restores those with no RID
+# under locked mode and fails NU1004. Packaging the desktop client would have broken the server's image
+# build. The gate that matters is the locked solution restore above, which is untouched.
+dotnet publish "$PROJECT" \
+    --configuration Release \
+    --runtime "$RUNTIME" \
+    --self-contained true \
+    --output "$PUBLISH_DIR" \
+    -p:RestoreLockedMode=false \
+    || stop_with 'Publish failed.'
+
+# An unlocked restore rewrites the lock files it walked. Left there, the next commit would carry exactly
+# the change that breaks the image build. Safe to do bluntly because this script refuses to run on a dirty
+# tree, so anything modified here is its own.
+git checkout -- '*packages.lock.json' || stop_with 'Could not restore the lock files after publishing.'
+
+# Checked rather than assumed. A publish directory without Velopack.dll would pack into an installer for an
+# application that never checks for updates — which looks completely normal until the next release goes out
+# and nobody receives it.
+for required in DodoSSH Velopack.dll; do
+    [ -e "$PUBLISH_DIR/$required" ] || stop_with "$required is missing from $PUBLISH_DIR."
+done
+
+echo "  $(du -sh "$PUBLISH_DIR" | cut -f1) in $(find "$PUBLISH_DIR" -type f | wc -l | tr -d ' ') files"
+
+# ---- Signing the native libraries, before vpk signs anything ------------------------------------------
+
+# ◆ THIS LOOP IS WHY NOTARIZATION SUCCEEDS, AND IT LOOKS REDUNDANT.
+#
+# vpk signs the finished bundle itself, with `codesign -f -v --timestamp --options runtime --entitlements
+# <file> --deep`, and --deep is documented by Apple as the wrong way to sign nested code. Apple's guidance
+# is inside-out: sign each nested binary first, then the bundle around it. --deep does the reverse in one
+# pass and applies the outer entitlements to everything it touches.
+#
+# In practice --deep alone is where the failure recorded in docs/platform-flags.md comes from — a
+# notarization rejection that does not name the offending file, on a submission that took its time getting
+# there. Signing each dylib properly first means vpk's pass has nothing left to get wrong, and re-signing
+# an already correctly signed binary with -f is a no-op in effect.
+#
+# No --entitlements here, and that is the difference that matters. Entitlements belong on the main
+# executable; a dylib carrying allow-jit is at best meaningless and at worst a rejection.
+write_step 'Signing native libraries'
+
+# createdump is a Mach-O executable the runtime ships and it is signed like the libraries: a nested
+# executable that is not signed fails notarization exactly as an unsigned dylib does, and it is the one
+# people forget because it has no extension to grep for.
+NATIVE_COUNT=0
+while IFS= read -r -d '' binary; do
+    codesign --force --verbose=0 --timestamp --options runtime \
+        --sign "$DODOSSH_SIGN_APP_IDENTITY" "$binary" \
+        || stop_with "codesign failed on $binary"
+    NATIVE_COUNT=$((NATIVE_COUNT + 1))
+done < <(find "$PUBLISH_DIR" \( -name '*.dylib' -o -name 'createdump' \) -type f -print0)
+
+[ "$NATIVE_COUNT" -gt 0 ] || stop_with "No native binaries found under $PUBLISH_DIR, which cannot be right for a self-contained publish."
+echo "  signed $NATIVE_COUNT native binaries"
+
+# ---- The bundle's Info.plist --------------------------------------------------------------------------
+
+# Rendered rather than committed, because vpk copies a custom plist verbatim and substitutes nothing —
+# so a committed one would carry whatever version it was written with into every release afterwards.
+# See the header of build/macos/Info.plist.template.
+write_step "Rendering Info.plist for $PLIST_VERSION"
+RENDERED_PLIST="$(mktemp -t dodossh-plist)"
+trap 'rm -f "$RENDERED_PLIST"' EXIT
+
+sed "s/@VERSION@/$PLIST_VERSION/g" "$PLIST_TEMPLATE" > "$RENDERED_PLIST"
+
+# The placeholder is the whole mechanism, so its absence is checked rather than hoped for. A template
+# somebody edited into a literal version would otherwise sail through and pin every future release to it.
+grep -q '@VERSION@' "$PLIST_TEMPLATE" || stop_with "$PLIST_TEMPLATE has no @VERSION@ placeholder left in it."
+! grep -q '@VERSION@' "$RENDERED_PLIST" || stop_with 'Substitution into the rendered Info.plist did not take.'
+
+mkdir -p "$RELEASES_DIR"
+
+# The previous release, so a delta can be built against it. Tolerated when it finds nothing: the first
+# macOS release has no predecessor, and a hard failure here would make cutting it impossible.
+write_step 'Fetching the previous release, for deltas'
+if ! dotnet vpk download gitea --repoUrl "$REPO_URL" --outputDir "$RELEASES_DIR" --channel "$CHANNEL"; then
+    echo '  Nothing came down. This package will be full-only, which is right for a first release.'
+fi
+
+# ---- Pack, sign, notarize, staple ---------------------------------------------------------------------
+
+# One command does the rest, and it is worth knowing what it is doing on your behalf, because the slow
+# part is not local: it builds the .app from the published files, signs it with the Developer ID
+# certificate and the entitlements below, submits it to Apple with `xcrun notarytool submit --wait`,
+# staples the resulting ticket to the package, and then builds the .pkg installer and the release index.
+#
+# The notarization wait is the reason this step can take a quarter of an hour and occasionally much
+# longer — it is a queue at Apple, not a computation here, and vpk's own message says so.
+#
+# --signInstallIdentity is a different certificate from --signAppIdentity, and the pair is not
+# interchangeable: "Developer ID Application" signs the bundle, "Developer ID Installer" signs the .pkg.
+# Passing one where the other belongs fails with a message about an identity that cannot be found, which
+# reads like a keychain problem rather than like the wrong certificate.
+write_step 'Packing, signing and notarizing (the notarization wait is Apple queueing, not this machine)'
+
+dotnet vpk pack \
+    --packId "$PACK_ID" \
+    --packVersion "$VERSION" \
+    --packDir "$PUBLISH_DIR" \
+    --packTitle "$PACK_TITLE" \
+    --packAuthors "$PACK_AUTHORS" \
+    --mainExe 'DodoSSH' \
+    --icon "$ICON" \
+    --plist "$RENDERED_PLIST" \
+    --entitlements "$ENTITLEMENTS" \
+    --signAppIdentity "$DODOSSH_SIGN_APP_IDENTITY" \
+    --signInstallIdentity "$DODOSSH_SIGN_INSTALL_IDENTITY" \
+    --notaryProfile "$DODOSSH_NOTARY_PROFILE" \
+    --runtime "$RUNTIME" \
+    --channel "$CHANNEL" \
+    --outputDir "$RELEASES_DIR" \
+    || stop_with 'vpk pack failed.'
+
+# ---- Did the notarization actually take? --------------------------------------------------------------
+
+# Asked rather than assumed, and this is the check worth having above all the others. A package whose
+# ticket did not staple is indistinguishable from a good one on the machine that built it — the Mac that
+# signed something trusts it locally — and reveals itself only on somebody else's machine, as a refusal
+# to open at all. spctl assesses it the way Gatekeeper will on a machine that has never seen this
+# certificate.
+write_step 'Verifying the notarization the way another Mac will'
+
+PKG="$(ls -t "$RELEASES_DIR"/*.pkg 2>/dev/null | head -n 1)"
+[ -n "$PKG" ] || stop_with 'vpk pack reported success but produced no .pkg.'
+
+if ! spctl --assess --type install --verbose=4 "$PKG"; then
+    stop_with "Gatekeeper rejects $PKG. It is signed but the notarization ticket is missing or stale; do not upload it."
+fi
+
+xcrun stapler validate "$PKG" || stop_with "The notarization ticket is not stapled to $PKG."
+
+write_step 'Built, notarized, and deliberately not uploaded'
+
+ls -lh "$RELEASES_DIR" | tail -n +2
+
+cat <<EOF
+
+Next:
+  1. Install the .pkg above and walk Phase 18 of docs/manual-checks.md.
+  2. Then: bash scripts/release-macos.sh --upload
+EOF

src/DodoSSH.Client.Android/App.axaml.cs

@@ -90,32 +90,61 @@ public sealed partial class DodoSshApp : Avalonia.Application
 
         shell.DataContext = viewModel;
 
-        // Difference 2: the foreground service, which is what makes TerminalWorkspace's promise — that a
-        // shell outlives a vault lock — true on a platform that stops backgrounded processes.
-        //
-        // The transfer count is real now that the document picker gives this head a way to start one, and
-        // it is the half that matters most here: a shell survives backgrounding because somebody is looking
-        // at it, and an upload has to survive precisely when nobody is — the screen is off and the phone is
-        // in a pocket. Queued counts as active, so putting five files in the queue and locking the phone
-        // moves five files.
-        //
-        // A local rather than a field, matching the desktop head: an Avalonia Application has no disposal
-        // hook, so a field holding a disposable would have nowhere honest to release it. It stays alive
-        // because it is subscribed to the workspace, which lives as long as the process.
-        var keepAlive = new SessionKeepAlive(
-            workspace,
-            activeTransfers: () => viewModel.Transfers.ActiveTransfers);
-
-        // The other end of the same wire: the workspace announces its own sessions ending, and the queue
-        // announces transfers appearing and finishing. Without this the notification would come up when an
-        // upload started and stay up after it finished, which is the failure this class exists to prevent.
-        viewModel.Transfers.ActivityChanged += (_, _) => keepAlive.Refresh();
-
-        keepAlive.Refresh();
+        // Difference 2, wired up in its own method purely for length — see ComposeKeepAlive for what it
+        // does and why.
+        ComposeKeepAlive(workspace, viewModel);
 
         return shell;
     }
 
+    /// <summary>
+    /// Wires up the foreground service that makes <c>TerminalWorkspace</c>'s promise — that a shell outlives
+    /// a vault lock — true on a platform that stops backgrounded processes.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// Split out of <see cref="Compose"/> for length rather than for reuse; there is exactly one caller.
+    /// </para>
+    /// <para>
+    /// The transfer count is real now that the document picker gives this head a way to start one, and it
+    /// matters exactly when nobody is looking: a shell survives backgrounding because somebody opened it,
+    /// and an upload has to survive precisely when nobody is — the screen is off and the phone is in a
+    /// pocket. Queued counts as active, so putting five files in the queue and locking the phone moves five
+    /// files. <c>holdsFileSession</c> covers the third case a count alone cannot: a host connected on the
+    /// Files screen with no transfer moving is still a live SFTP session that backgrounding would sever, and
+    /// <c>TransfersViewModel.HasLiveFileSession</c> is the existing fact — <c>IsConnected</c> with a real
+    /// cipher, which a bucket never has — that answers whether one is open.
+    /// </para>
+    /// <para>
+    /// <c>keepAlive</c> is a local rather than a field, matching the desktop head: an Avalonia Application
+    /// has no disposal hook, so a field holding a disposable would have nowhere honest to release it. It
+    /// stays alive because it is subscribed to the workspace, which lives as long as the process.
+    /// </para>
+    /// </remarks>
+    private static void ComposeKeepAlive(TerminalWorkspace workspace, MainWindowViewModel viewModel)
+    {
+        var keepAlive = new SessionKeepAlive(
+            workspace,
+            activeTransfers: () => viewModel.Transfers.ActiveTransfers,
+            holdsFileSession: () => viewModel.Transfers.HasLiveFileSession);
+
+        // The other end of the same wire: the workspace announces its own sessions ending, and the queue
+        // announces transfers appearing and finishing, and a Files session connecting or disconnecting.
+        // Without this the notification would come up when an upload started and stay up after it
+        // finished, which is the failure this class exists to prevent.
+        viewModel.Transfers.ActivityChanged += (_, _) => keepAlive.Refresh();
+
+        // The half that was missing until now: a shell opening. SessionKeepAlive already heard the
+        // workspace announce a session ending, but nothing announced the opposite — a user who opened a
+        // shell and backgrounded the app had no foreground service at all, because the only wire in was the
+        // one for taking it down. TerminalSessionOpened is that other half, forwarded from
+        // VaultViewModel.SessionOpened, and without this line the service could never come up for a shell
+        // in the first place, which was precisely the promise this whole arrangement exists to keep.
+        viewModel.TerminalSessionOpened += (_, _) => keepAlive.Refresh();
+
+        keepAlive.Refresh();
+    }
+
     /// <summary>
     /// Writing to this phone's clipboard.
     /// </summary>

src/DodoSSH.Client.Android/Platform/SessionForegroundService.cs

@@ -6,7 +6,7 @@ using global::Android.OS;
 namespace DodoSSH.Client.Android.Platform;
 
 /// <summary>
-/// Keeps the process alive for as long as a shell or a transfer is live.
+/// Keeps the process alive for as long as a shell, a transfer, or a connected Files session is live.
 /// </summary>
 /// <remarks>
 /// <para>
@@ -41,6 +41,26 @@ internal sealed class SessionForegroundService : Service
     private const string ChannelId = "dodossh.sessions";
     private const int NotificationId = 1;
 
+    // API 33+ requires the request to name a code the RequestPermissionsResult callback would be handed
+    // back — 1 is fine because this head never implements that callback at all, see
+    // RequestNotificationPermission's own remark for why a result is not worth listening for.
+    private const int NotificationPermissionRequestCode = 1;
+
+    /// <summary>
+    /// Whether <see cref="OnStartCommand"/> has run for this process without a matching
+    /// <see cref="OnDestroy"/> since — i.e. whether Android currently considers this service foregrounded.
+    /// </summary>
+    /// <remarks>
+    /// Volatile because <see cref="Reconcile"/> can run on whatever thread called
+    /// <see cref="SessionKeepAlive.Refresh"/>, while this is set from the binder thread Android delivers
+    /// service lifecycle callbacks on — two threads with no other synchronisation between them, and a stale
+    /// read here is the difference between updating a notification in place and calling
+    /// <c>StartForegroundService</c> on a service that is already running, which is what defect 3 was.
+    /// </remarks>
+    private static volatile bool running;
+
+    private static bool notificationPermissionRequested;
+
     /// <remarks>
     /// A bound service would tie the sessions' lifetime to a binding, which is the opposite of what is
     /// wanted here: the point is that they outlive whatever the user does with the interface.
@@ -49,7 +69,9 @@ internal sealed class SessionForegroundService : Service
 
     public override StartCommandResult OnStartCommand(Intent? intent, StartCommandFlags flags, int startId)
     {
-        StartForeground(NotificationId, BuildNotification(intent?.GetStringExtra("summary") ?? "Working"));
+        running = true;
+
+        StartForeground(NotificationId, BuildNotification(this, intent?.GetStringExtra("summary") ?? "Working"));
 
         // NotSticky: if Android does kill this process, the SSH connections died with it and there is
         // nothing to resume. Restarting the service would produce a notification claiming sessions that no
@@ -58,14 +80,36 @@ internal sealed class SessionForegroundService : Service
         return StartCommandResult.NotSticky;
     }
 
+    /// <inheritdoc />
     /// <remarks>
+    /// The other half of <see cref="running"/>. Android calls this whether the service stopped itself or
+    /// was stopped from outside — <see cref="Reconcile"/>'s down case calls <c>StopService</c> rather than
+    /// clearing the flag directly, so this override is the one place that actually knows the service has
+    /// gone, matching how <see cref="OnStartCommand"/> is the one place that knows it has come up.
+    /// </remarks>
+    public override void OnDestroy()
+    {
+        running = false;
+
+        base.OnDestroy();
+    }
+
+    /// <remarks>
+    /// <para>
     /// Low importance on purpose. This notification is a receipt, not an alert — it exists because Android
     /// requires one, and because the user is entitled to know the app is holding connections open. Making
     /// it buzz would be a notification about nothing having happened.
+    /// </para>
+    /// <para>
+    /// Static, taking the <see cref="Context"/> it needs rather than reading <c>this</c>: the instance path
+    /// through <see cref="OnStartCommand"/> passes the service itself, and the in-place update path through
+    /// <see cref="Reconcile"/> has no service instance at all — only <see cref="PhoneEnvironment.Require"/>
+    /// — because posting to an already-running notification never touches the service's own lifecycle.
+    /// </para>
     /// </remarks>
-    private Notification BuildNotification(string summary)
+    private static Notification BuildNotification(Context context, string summary)
     {
-        var manager = (NotificationManager)GetSystemService(NotificationService)!;
+        var manager = (NotificationManager)context.GetSystemService(Context.NotificationService)!;
 
         if (OperatingSystem.IsAndroidVersionAtLeast(26))
         {
@@ -79,12 +123,12 @@ internal sealed class SessionForegroundService : Service
         }
 
         var reopen = PendingIntent.GetActivity(
-            this,
+            context,
             0,
-            new Intent(this, typeof(MainActivity)).SetFlags(ActivityFlags.SingleTop),
+            new Intent(context, typeof(MainActivity)).SetFlags(ActivityFlags.SingleTop),
             PendingIntentFlags.Immutable | PendingIntentFlags.UpdateCurrent);
 
-        return new Notification.Builder(this, ChannelId)
+        return new Notification.Builder(context, ChannelId)
             .SetContentTitle("DodoSSH")
             .SetContentText(summary)
             .SetSmallIcon(global::Android.Resource.Drawable.IcDialogInfo)
@@ -93,37 +137,138 @@ internal sealed class SessionForegroundService : Service
             .Build();
     }
 
-    /// <summary>Starts or stops the service to match what is actually running.</summary>
+    /// <summary>Starts, stops, or refreshes the service's notification to match what is actually running.</summary>
     /// <param name="liveSessions">Shells with a live channel behind them.</param>
     /// <param name="activeTransfers">Transfers still moving bytes.</param>
-    public static void Reconcile(int liveSessions, int activeTransfers)
+    /// <param name="holdsFileSession">Whether the Files screen holds a live, idle SFTP connection.</param>
+    /// <remarks>
+    /// Three outcomes, not the two a plain start-or-stop would have. Nothing live stops the service, as
+    /// always. Something live and the service not yet running starts it. Something live and the service
+    /// already running is the case that used to call <c>StartForegroundService</c> a second time, which on
+    /// API 31+ throws <c>ForegroundServiceStartNotAllowedException</c> the instant the app is backgrounded
+    /// — a transfer finishing in the pocket, one of two shells dying — crashing the process and taking the
+    /// remaining connections with it. That case — running, and the app backgrounded — now only posts a
+    /// fresh notification through the <see cref="NotificationManager"/> already holding the channel open,
+    /// which needs no foreground-start permission at all. A foregrounded refresh still prefers a real
+    /// start even over a service that looks like it is running; the inline remark below is why.
+    /// </remarks>
+    public static void Reconcile(int liveSessions, int activeTransfers, bool holdsFileSession)
     {
         var context = PhoneEnvironment.Require();
-        var intent = new Intent(context, typeof(SessionForegroundService));
 
-        if (liveSessions == 0 && activeTransfers == 0)
+        if (liveSessions == 0 && activeTransfers == 0 && !holdsFileSession)
         {
-            context.StopService(intent);
+            context.StopService(new Intent(context, typeof(SessionForegroundService)));
 
             return;
         }
 
         // The summary says what is actually held, counted rather than generic — the same principle the
         // delete confirmations follow. "DodoSSH is running" would tell the user nothing they could act on.
-        intent.PutExtra("summary", Summarise(liveSessions, activeTransfers));
+        var summary = Summarise(liveSessions, activeTransfers, holdsFileSession);
 
-        context.StartForegroundService(intent);
+        // In-place only while backgrounded, where it is the only legal move. Foregrounded, a real start is
+        // always allowed and is preferred even when `running` says the service is up: a disconnect followed
+        // by a quick reconnect can land here while the StopService just issued is still in flight, and
+        // posting to that dying service's notification would leave an orphan receipt over an unprotected
+        // process — restarting instead makes the flag's small lag harmless. A start on a service that
+        // really is running only re-delivers OnStartCommand, whose StartForeground updates the same
+        // notification anyway. CurrentActivity is the foreground signal: set on resume, cleared on pause.
+        if (running && PhoneEnvironment.CurrentActivity is null)
+        {
+            var manager = (NotificationManager)context.GetSystemService(Context.NotificationService)!;
+            manager.Notify(NotificationId, BuildNotification(context, summary));
+
+            return;
+        }
+
+        RequestNotificationPermission(context);
+        Start(context, summary);
     }
 
-    private static string Summarise(int liveSessions, int activeTransfers)
+    /// <remarks>
+    /// Split out of <see cref="Reconcile"/> for the try/catch alone, which needs its own remark and would
+    /// otherwise crowd the three-way branch above it.
+    /// </remarks>
+    private static void Start(Context context, string summary)
     {
-        var parts = new List<string>(2);
+        var intent = new Intent(context, typeof(SessionForegroundService));
+        intent.PutExtra("summary", summary);
+
+        try
+        {
+            context.StartForegroundService(intent);
+        }
+        catch (Java.Lang.IllegalStateException)
+        {
+            // ForegroundServiceStartNotAllowedException (API 31+) derives from this, and reaching it here
+            // means a start-worthy transition — a shell opening, a Files connection completing, the first
+            // transfer landing in an empty queue — happened while the app was backgrounded, which is
+            // precisely when Android refuses a new foreground start. There is no retry that helps: by the
+            // time this catch runs, the moment such a start would have been allowed has already passed.
+            // Swallowing it is the honest choice and not just the available one — letting the exception
+            // propagate would crash the process and drop the very shells and transfers this service exists
+            // to keep alive. A process that keeps running unprotected outlives one that does not run at all.
+        }
+    }
+
+    /// <summary>
+    /// Asks for the receipt notification's own permission, the first time this process actually has
+    /// something to show rather than at launch.
+    /// </summary>
+    /// <remarks>
+    /// At most once per process, via <see cref="notificationPermissionRequested"/> — not to work around a
+    /// platform limit, since Android already refuses to show the dialogue twice, but because a second call
+    /// to <c>RequestPermissions</c> after the first is still pending is its own kind of noise. No result is
+    /// read back: there is nothing this class would do differently for a grant versus a refusal, so a
+    /// callback would exist only to be empty. What refusal costs is stated rather than hidden — the
+    /// notification stays invisible — and what it does not cost is the point: the service still starts,
+    /// still holds the process in the foreground, and the shells and transfers it protects are exactly as
+    /// safe as if the user had said yes. See the manifest's own comment on this permission.
+    /// </remarks>
+    private static void RequestNotificationPermission(Context context)
+    {
+        // Isolated as its own guard clause rather than folded into the compound condition below: the
+        // platform-compatibility analyzer only recognises a version check as guarding what follows when it
+        // is the sole condition of its own early return, and PostNotifications is annotated API 33+.
+        if (!OperatingSystem.IsAndroidVersionAtLeast(33))
+        {
+            return;
+        }
+
+        // Read into a local rather than referenced again inside the lambda below: the guard clause above
+        // covers a direct call in this method's own body, but the platform-compatibility analyzer treats a
+        // lambda as reachable from anywhere and will not extend the guard across that boundary. Capturing
+        // the already-validated string sidesteps the false positive without weakening the actual check.
+        var postNotifications = global::Android.Manifest.Permission.PostNotifications;
+
+        if (notificationPermissionRequested
+            || PhoneEnvironment.CurrentActivity is not { } activity
+            || context.CheckSelfPermission(postNotifications) == Permission.Granted)
+        {
+            return;
+        }
+
+        notificationPermissionRequested = true;
+
+        activity.RunOnUiThread(() =>
+            activity.RequestPermissions([postNotifications], NotificationPermissionRequestCode));
+    }
+
+    private static string Summarise(int liveSessions, int activeTransfers, bool holdsFileSession)
+    {
+        var parts = new List<string>(3);
 
         if (liveSessions > 0)
         {
             parts.Add(liveSessions == 1 ? "1 shell connected" : $"{liveSessions} shells connected");
         }
 
+        if (holdsFileSession)
+        {
+            parts.Add("Files connected");
+        }
+
         if (activeTransfers > 0)
         {
             parts.Add(activeTransfers == 1 ? "1 transfer running" : $"{activeTransfers} transfers running");

src/DodoSSH.Client.Android/Platform/SessionKeepAlive.cs

@@ -17,37 +17,53 @@ namespace DodoSSH.Client.Android.Platform;
 /// tally kept here. It already knows that a session whose shell exited half an hour ago is not live, which
 /// a counter incremented on open and decremented on close would not.
 /// </para>
+/// <para>
+/// Three facts feed <see cref="SessionForegroundService.Reconcile"/>, not one: live shells, moving
+/// transfers, and an idle-but-connected Files session. The last of those used to be missing entirely —
+/// a shell survives backgrounding because somebody opened it, but a Files connection with nothing moving
+/// looked, to this class, exactly like nothing being open at all. <c>holdsFileSession</c> below is that
+/// gap closed, read the same way the other two facts are: asked, not cached.
+/// </para>
 /// </remarks>
 internal sealed class SessionKeepAlive : IDisposable
 {
     private readonly TerminalWorkspace workspace;
     private readonly Func<int> activeTransfers;
+    private readonly Func<bool> holdsFileSession;
 
     /// <param name="workspace">The live shells.</param>
     /// <param name="activeTransfers">
-    /// How many transfers are moving bytes. A delegate rather than a queue, because file transfer is out
-    /// of this head's first scope — see the decision in docs/android-port.md — and this is the seam it
-    /// will arrive through rather than a dependency taken before there is anything to depend on.
+    /// How many transfers are moving bytes. A delegate rather than a queue, because ownership of the
+    /// transfer queue stays with <c>TransfersViewModel</c> — this class only ever asks it a question.
     /// </param>
-    public SessionKeepAlive(TerminalWorkspace workspace, Func<int> activeTransfers)
+    /// <param name="holdsFileSession">
+    /// Whether the Files screen holds a live SFTP connection with nothing moving on it — the idle-but-
+    /// connected case a transfer count alone would miss. See <c>TransfersViewModel.HasLiveFileSession</c>.
+    /// </param>
+    public SessionKeepAlive(TerminalWorkspace workspace, Func<int> activeTransfers, Func<bool> holdsFileSession)
     {
         this.workspace = workspace;
         this.activeTransfers = activeTransfers;
+        this.holdsFileSession = holdsFileSession;
 
-        // Raised on whatever thread the pump unwound on, which is fine: starting and stopping a service is
-        // a binder call and needs no particular thread. Nothing here touches the interface.
+        // Raised on whatever thread the workspace announced from — a continuation of the ended run, or the
+        // closer's own — which is fine: starting and stopping a service is a binder call and needs no
+        // particular thread. Nothing here touches the interface. That the announcement waits for the run to
+        // actually complete, and comes for deliberate closes too, is what makes reading LiveSessionCount
+        // from it honest — the event's own remark carries the stuck notification that taught us both.
         workspace.SessionEnded += OnSessionEnded;
     }
 
     /// <summary>Re-reads the counts and starts or stops the service to match.</summary>
     /// <remarks>
-    /// Called after anything that could change either count — opening a shell, closing a tab, a transfer
-    /// finishing. Calling it when nothing changed is free: reconciling to the state it is already in is
-    /// either a redundant <c>startForegroundService</c> on a running service or a <c>stopService</c> on a
-    /// stopped one, and Android treats both as no-ops.
+    /// Called after anything that could change any of the three facts — opening a shell, closing a tab, a
+    /// transfer finishing, a Files connection opening or closing. Calling it when nothing changed is free:
+    /// reconciling to the state it is already in is either a redundant <c>startForegroundService</c> — or,
+    /// now, a redundant notification post — on a running service, or a <c>stopService</c> on a stopped one,
+    /// and Android treats all of those as no-ops.
     /// </remarks>
     public void Refresh() =>
-        SessionForegroundService.Reconcile(workspace.LiveSessionCount, activeTransfers());
+        SessionForegroundService.Reconcile(workspace.LiveSessionCount, activeTransfers(), holdsFileSession());
 
     /// <inheritdoc />
     public void Dispose()
@@ -56,7 +72,7 @@ internal sealed class SessionKeepAlive : IDisposable
 
         // The notification goes with the composition root. Leaving it up over a process that is shutting
         // down is how an SSH client acquires a reputation for a notification you cannot get rid of.
-        SessionForegroundService.Reconcile(0, 0);
+        SessionForegroundService.Reconcile(0, 0, false);
     }
 
     private void OnSessionEnded(object? sender, TerminalSessionEndedEventArgs e) => Refresh();

src/DodoSSH.Client.Android/Platform/TerminalFocus.cs

@@ -0,0 +1,79 @@
+using global::Android.Views;
+
+namespace DodoSSH.Client.Android.Platform;
+
+/// <summary>
+/// Hands native focus back to the terminal's WebView after Avalonia chrome took it.
+/// </summary>
+/// <remarks>
+/// <para>
+/// <b>The sibling of <see cref="SoftKeyboard"/>, and it exists for the same reason that one does:</b> the
+/// keyboard over a terminal belongs to the WebView's own native view, which Avalonia's focus manager does
+/// not own. The accessory row's keys are already <c>Focusable=false</c> — see TerminalScreen — so Avalonia's
+/// idea of focus never leaves the terminal when one is tapped. What still moves is <em>Android's</em>:
+/// <c>AvaloniaView.DispatchTouchEvent</c> (decompiled from Avalonia.Android 12.1.1) ends every handled
+/// touch — DOWN and UP alike — with a <c>RequestFocus()</c> for Avalonia's own view. The WebView's input
+/// connection dies with its focus, the keyboard swaps to the layout it shows an editor that takes no text,
+/// and the inset churn that follows can leave it sitting on top of the very row that was tapped.
+/// </para>
+/// <para>
+/// <b>Posted, not called — the posting is the fix's second attempt, and the first one's failure is why.</b>
+/// The first version called <c>RequestFocus()</c> from the keys' own Click handlers, which fire
+/// <em>inside</em> the UP event's dispatch — and the platform's own request runs <em>after</em> dispatch
+/// returns, so it undid ours a few microseconds later and the terminal stayed unfocused. A posted runnable
+/// runs on the next main-looper message, after the platform has taken its turn, so ours is the request that
+/// sticks. The focus check lives inside the posted runnable for the same reason: the answer at call time is
+/// about to be made stale by the very mechanism this exists to counter.
+/// </para>
+/// <para>
+/// The page inside the WebView never noticed any of this — its own DOM focus never moved — so regaining
+/// native focus re-establishes the same input connection and the keyboard settles back to what it was.
+/// </para>
+/// <para>
+/// Found by walking the decor view rather than asked of the <c>NativeWebView</c> control, because the
+/// control does not expose its platform child and this application only ever has the one WebView — the
+/// walk's first match is necessarily the terminal. Every step is allowed to be absent, exactly as
+/// <see cref="SoftKeyboard.Hide"/>'s are: no activity while backgrounded, no WebView while the terminal
+/// surface has never been shown, and nothing to do in either case.
+/// </para>
+/// </remarks>
+internal static class TerminalFocus
+{
+    public static void Return()
+    {
+        if (PhoneEnvironment.CurrentActivity?.Window?.DecorView is not ViewGroup decor)
+        {
+            return;
+        }
+
+        if (FindWebView(decor) is not { } webView)
+        {
+            return;
+        }
+
+        webView.Post(() =>
+        {
+            if (!webView.IsFocused)
+            {
+                webView.RequestFocus();
+            }
+        });
+    }
+
+    private static View? FindWebView(ViewGroup parent)
+    {
+        for (var i = 0; i < parent.ChildCount; i++)
+        {
+            switch (parent.GetChildAt(i))
+            {
+                case global::Android.Webkit.WebView webView:
+                    return webView;
+
+                case ViewGroup child when FindWebView(child) is { } found:
+                    return found;
+            }
+        }
+
+        return null;
+    }
+}

src/DodoSSH.Client.Android/Properties/AndroidManifest.xml

@@ -12,7 +12,12 @@
   <uses-permission android:name="android.permission.FOREGROUND_SERVICE" />
   <uses-permission android:name="android.permission.FOREGROUND_SERVICE_DATA_SYNC" />
 
-  <!-- The service's persistent notification. Runtime-requested on API 33+, and refusal is survivable. -->
+  <!--
+    The service's persistent notification. Requested on API 33+ from SessionForegroundService.Reconcile,
+    the first time in this process there is actually something to show — not at launch, where the ask
+    would justify nothing on screen yet. Refusal is survivable: the service still starts and still holds
+    the process in the foreground either way, so a "no" costs the notification and nothing else.
+  -->
   <uses-permission android:name="android.permission.POST_NOTIFICATIONS" />
 
   <!-- Releases the device key. See AndroidDeviceKeyStore. -->

src/DodoSSH.Client.Android/Theme/Phone.axaml

@@ -116,11 +116,27 @@
     <Setter Property="FontWeight" Value="SemiBold" />
   </Style>
 
-  <!-- A row in a list: the whole row is the target, and it is 54 tall because a thumb is not a mouse. -->
+  <!--
+    A row in a list: the whole row is the target, and it is 54 tall because a thumb is not a mouse.
+
+    ◆ VerticalContentAlignment, because that height is the whole point of this class and Avalonia's default
+    for content alignment is Stretch — so the content presenter stretched the caption to the full row and a
+    TextBlock draws its line at the TOP of what it is given. Most rows here never showed it, having a
+    StackPanel or a Grid of already-centred children in them, which is what made the four that did look
+    like four unrelated mistakes: the breadcrumb chips and the up-one-directory button on FilesScreen, and
+    TerminalScreen's CLOSE THIS TAB, each a bare TextBlock in a row 36 or 44 tall with no vertical padding.
+    Measured at those numbers, the caption sat flush against the top edge with 21 to 33 pixels below it.
+
+    The desktop head's App.axaml carries the same setter on its own shapes for the same reason, and excludes
+    two of them — see the remark on Button.ghost there. Nothing is excluded here: no row's content depends
+    on being stretched, there being no full-height strip inside any of the thirty-three, and the Grids that
+    stop filling hold only children that already centre themselves, so they land where they always did.
+  -->
   <Style Selector="Button.row">
     <Setter Property="MinHeight" Value="54" />
     <Setter Property="HorizontalAlignment" Value="Stretch" />
     <Setter Property="HorizontalContentAlignment" Value="Stretch" />
+    <Setter Property="VerticalContentAlignment" Value="Center" />
     <Setter Property="Background" Value="Transparent" />
     <Setter Property="BorderThickness" Value="0" />
     <Setter Property="CornerRadius" Value="10" />

src/DodoSSH.Client.Android/Views/FilesScreen.axaml

@@ -48,7 +48,7 @@
     the refusal has no continue button here either.
   -->
 
-  <Grid RowDefinitions="Auto,Auto,Auto,Auto,*,Auto">
+  <Grid RowDefinitions="Auto,Auto,Auto,Auto,Auto,*,Auto">
 
     <!-- ============ header ============ -->
     <Grid Grid.Row="0" ColumnDefinitions="Auto,Auto,*,Auto" Height="56" Margin="8,0">
@@ -199,8 +199,48 @@
       </StackPanel>
     </ScrollViewer>
 
+    <!-- ============ pins ============ -->
+    <!--
+      ◆ The phone's answer to the desktop's QUICK ACCESS sidebar — the same PinnedPathList this host was
+      connected with, drawn where this head actually browses files rather than beside a terminal it has no
+      strip for. See TransfersViewModel.ConnectedPinnedPaths for why this is a snapshot taken at connect
+      rather than a live follow of the host row: a pin added or removed mid-session shows up here on the
+      next connect, not while this one is still open.
+
+      Its own row rather than folded into the breadcrumb above: the two scroll independently and mean
+      different things — the breadcrumb is where this pane is, the chips are places it can jump to — and a
+      shared row would make ↑ look like one more pin among several.
+
+      Gone rather than empty when nothing is pinned or nothing is connected, which HasConnectedPins already
+      answers: an empty scrolling strip under the breadcrumb would read as a loading row rather than as
+      "this host has nothing pinned".
+    -->
+    <ScrollViewer Grid.Row="4" HorizontalScrollBarVisibility="Auto" VerticalScrollBarVisibility="Disabled"
+                  IsVisible="{Binding HasConnectedPins}" Margin="0,0,0,4">
+      <ItemsControl ItemsSource="{Binding ConnectedPinnedPaths}">
+        <ItemsControl.ItemsPanel>
+          <ItemsPanelTemplate><StackPanel Orientation="Horizontal" Spacing="6" Margin="14,0" /></ItemsPanelTemplate>
+        </ItemsControl.ItemsPanel>
+        <ItemsControl.ItemTemplate>
+          <DataTemplate x:DataType="x:String">
+            <!--
+              Classes="row" at its default 44dp MinHeight rather than the breadcrumb's own 36 — these chips
+              are the destination, not the trail behind it, and a target worth a deliberate tap gets the
+              full touch floor this head holds everything else to. Not trimmed: a chip scrolls sideways with
+              the row rather than being squeezed to fit it, so the full path is always what a tap commits to.
+            -->
+            <Button Classes="row" MinHeight="44" Padding="11,0" CornerRadius="10"
+                    Command="{Binding $parent[views:FilesScreen].((vm:TransfersViewModel)DataContext).GoRemoteCommand}"
+                    CommandParameter="{Binding}">
+              <TextBlock Classes="mono" FontSize="12" Text="{Binding}" />
+            </Button>
+          </DataTemplate>
+        </ItemsControl.ItemTemplate>
+      </ItemsControl>
+    </ScrollViewer>
+
     <!-- ============ the listing ============ -->
-    <Panel Grid.Row="4">
+    <Panel Grid.Row="5">
 
       <TextBlock Classes="body" IsVisible="{Binding !IsConnected}" Margin="24,12"
                  VerticalAlignment="Top" Text="{Binding Status}" />
@@ -266,7 +306,7 @@
 
     <!-- ============ what to do with the chosen entry ============ -->
     <!-- DeepChrome rather than Chrome, since v5 — see the remark on PhoneShell's vault header. -->
-    <Border Grid.Row="5" IsVisible="{Binding IsConnected}" Background="{StaticResource DeepChrome}"
+    <Border Grid.Row="6" IsVisible="{Binding IsConnected}" Background="{StaticResource DeepChrome}"
             BorderBrush="{StaticResource Border}" BorderThickness="0,1,0,0" Padding="12,10">
       <StackPanel Spacing="9">
 

src/DodoSSH.Client.Android/Views/HostsScreen.axaml

@@ -1098,6 +1098,60 @@
                     Command="{Binding AddEditorTagCommand}" />
           </Grid>
 
+          <!--
+            ◆ QUICK ACCESS. Staged on the shared VaultViewModel.EditorPinnedPaths the way EditorTagChoices
+            stages tags above — populated when the editor opens, read back by BuildHost on SAVE, left alone
+            by CANCEL. Nothing here is phone-only: a pin added on this page is the same PinnedPathList entry
+            the desktop's drawer stages under the same field, so it syncs to every other client the moment
+            this vault does, exactly as a tag would.
+
+            No folder glyph in the row, unlike the desktop's — this head carries MonoFont and a heading font
+            (see Theme/Phone.axaml) and no icon font, so the desktop's &#xE2C7; would draw as a tofu box
+            rather than a folder here. The path text carries the row on its own, trimmed in a Grid's star
+            column rather than a StackPanel's — a StackPanel measures its children at infinity, so
+            TextTrimming never actually engages inside one; this is the same fix TransfersScreen's pane
+            headers needed for the same reason.
+
+            The remove button is Classes="icon" at its default 44x44, not the desktop's 22x22 close box —
+            see this file's touch-floor rule at the top of the class: a target sized for a mouse pointer is
+            not one a thumb can reliably land on.
+
+            See surface B, the Files screen's pin chips, for where these pins actually get used on this
+            head — the hint sentence below names it rather than the desktop's terminal strip, which does
+            not exist here.
+          -->
+          <TextBlock Classes="label" Text="QUICK ACCESS" Margin="0,4,0,0" />
+
+          <ItemsControl ItemsSource="{Binding EditorPinnedPaths}">
+            <ItemsControl.ItemsPanel>
+              <ItemsPanelTemplate><StackPanel Spacing="6" /></ItemsPanelTemplate>
+            </ItemsControl.ItemsPanel>
+            <ItemsControl.ItemTemplate>
+              <DataTemplate x:DataType="x:String">
+                <Border MinHeight="44" CornerRadius="10" Background="{StaticResource Field}"
+                        BorderBrush="{StaticResource BorderMid}" BorderThickness="1" Padding="12,0">
+                  <Grid ColumnDefinitions="*,Auto">
+                    <TextBlock Grid.Column="0" Classes="mono" FontSize="12.5" Text="{Binding}"
+                               VerticalAlignment="Center" TextTrimming="CharacterEllipsis" />
+                    <Button Grid.Column="1" Classes="icon" Content="✕"
+                            Command="{Binding $parent[ItemsControl].((vm:VaultViewModel)DataContext).RemoveEditorPinCommand}"
+                            CommandParameter="{Binding}" ToolTip.Tip="Unpins this path" />
+                  </Grid>
+                </Border>
+              </DataTemplate>
+            </ItemsControl.ItemTemplate>
+          </ItemsControl>
+
+          <Grid ColumnDefinitions="*,8,Auto">
+            <TextBox Grid.Column="0" Classes="field" Text="{Binding EditorNewPin}"
+                     PlaceholderText="/path/to/folder" />
+            <Button Grid.Column="2" Classes="secondary" Height="44" Width="72" Content="ADD"
+                    Command="{Binding AddEditorPinCommand}" />
+          </Grid>
+
+          <TextBlock Classes="body"
+                     Text="Pins show up as chips on the Files screen while you are connected to this host — tap one to jump straight there." />
+
           <!--
             ◆ The one control here that publishes something. Turning it on copies this host's address and
             port into plaintext columns the server can read, which is the single deliberate concession in
@@ -1117,6 +1171,18 @@
           <TextBlock Classes="body"
                      Text="Not built yet: this app always dials the host itself, so ticking this stores the address on the server and changes nothing about how the host is reached. When it works, the relay will dial on your behalf — which is why the address and port have to be stored in the clear. Everything else about the host stays encrypted either way, and a relayed host needs a port of its own rather than its group's." />
 
+          <!--
+            This page covers the whole screen while it is open, which is why it needs a Status line of its
+            own rather than relying on the one the list behind it draws at the top of HOSTS (see this file's
+            other Classes="detail" Text="{Binding Status}") — that TextBlock is off-screen for as long as
+            IsEditing is true. AddEditorTagCommand and AddEditorPinCommand both refuse silently into Status
+            rather than throwing, so without a line to draw it here the honesty rule would be broken: a
+            refusal that writes a message nothing on screen shows is the same, to whoever pressed ADD, as no
+            refusal at all.
+          -->
+          <TextBlock Classes="detail" Text="{Binding Status}" TextWrapping="Wrap"
+                     IsVisible="{Binding Status, Converter={x:Static StringConverters.IsNotNullOrEmpty}}" />
+
           <Grid ColumnDefinitions="*,8,*" Margin="0,6,0,0">
             <Button Grid.Column="0" Classes="primary" Height="44" Content="SAVE"
                     Command="{Binding SaveHostCommand}" />

src/DodoSSH.Client.Android/Views/TerminalScreen.axaml.cs

@@ -139,48 +139,18 @@ internal sealed partial class TerminalScreen : UserControl
     {
         var row = this.FindControl<StackPanel>("AccessoryKeys")!;
 
+        // Both halves of a press steal Android's own focus — the platform requests it for Avalonia's view
+        // after dispatching every handled touch, DOWN and UP alike; TerminalFocus carries the decompiled
+        // citation. Countered at the row rather than inside each key's Click, and for two reasons: Click
+        // only exists for the UP half, so a keyboard detached at DOWN would stay detached for the whole
+        // length of the press; and the Return is posted past the current dispatch, so its ordering against
+        // the key's own handler does not matter — which is what lets one pair of handlers cover ten keys.
+        row.AddHandler(PointerPressedEvent, (_, _) => TerminalFocus.Return(), RoutingStrategies.Tunnel);
+        row.AddHandler(PointerReleasedEvent, (_, _) => TerminalFocus.Return(), RoutingStrategies.Tunnel);
+
         foreach (var (label, bytes, latches) in Keys)
         {
-            var key = new Button
-            {
-                Content = new TextBlock
-                {
-                    Text = label,
-                    FontFamily = (FontFamily)Application.Current!.FindResource("MonoFont")!,
-                    FontSize = 11,
-                    HorizontalAlignment = HorizontalAlignment.Center,
-                    VerticalAlignment = VerticalAlignment.Center,
-                },
-
-                // 44 wide, and that is the number that matters: the design draws these flexed across the
-                // width, which at 360dp with ten keys is 32 pixels each — under every thumb-target
-                // guideline there is. The height came down with the row it sits in, from 38 to 30, and it
-                // costs nothing a width does: the keys are a single row with the terminal above and the
-                // system's gesture bar below, so there is no neighbour a short press can land on instead.
-                MinWidth = 44,
-                Height = 30,
-                Padding = new Thickness(10, 0),
-                CornerRadius = new CornerRadius(9),
-                Background = Palette("Panel"),
-                BorderBrush = Palette("BorderMid"),
-                BorderThickness = new Thickness(1),
-                Foreground = Palette("TextDim"),
-                HorizontalContentAlignment = HorizontalAlignment.Center,
-
-                // ◆ NOT FOCUSABLE, AND THAT IS THE WHOLE CONTROL RATHER THAN A DETAIL.
-                //
-                // These keys are an extension of the keyboard, not a place the keyboard should go. As
-                // ordinary buttons they took Avalonia's focus on tap, which takes it off the NativeWebView
-                // — and the package's own OnLostFocus then calls the adapter's ResignFocus(). So pressing
-                // Tab or an arrow handed the terminal one byte and took the keyboard away from it: the next
-                // thing typed on a hardware keyboard went nowhere, and the row went on working because its
-                // buttons are pressed rather than typed into, which is what makes it look like the terminal
-                // had died instead.
-                //
-                // Focusable=false is what a toolbar button is, and it means the focused element never
-                // changes: the WebView is still it, so nothing resigns and nothing has to be handed back.
-                Focusable = false,
-            };
+            var key = CreateKey(label);
 
             if (latches)
             {
@@ -196,6 +166,57 @@ internal sealed partial class TerminalScreen : UserControl
         }
     }
 
+    /// <summary>One key of the accessory row, before its click is wired.</summary>
+    /// <remarks>Split from <see cref="BuildAccessoryRow"/> for length rather than for reuse.</remarks>
+    private static Button CreateKey(string label) =>
+        new()
+        {
+            Content = new TextBlock
+            {
+                Text = label,
+                FontFamily = (FontFamily)Application.Current!.FindResource("MonoFont")!,
+                FontSize = 11,
+                HorizontalAlignment = HorizontalAlignment.Center,
+                VerticalAlignment = VerticalAlignment.Center,
+            },
+
+            // 44 wide, and that is the number that matters: the design draws these flexed across the
+            // width, which at 360dp with ten keys is 32 pixels each — under every thumb-target
+            // guideline there is. The height came down with the row it sits in, from 38 to 30, and it
+            // costs nothing a width does: the keys are a single row with the terminal above and the
+            // system's gesture bar below, so there is no neighbour a short press can land on instead.
+            MinWidth = 44,
+            Height = 30,
+            Padding = new Thickness(10, 0),
+            CornerRadius = new CornerRadius(9),
+            Background = Palette("Panel"),
+            BorderBrush = Palette("BorderMid"),
+            BorderThickness = new Thickness(1),
+            Foreground = Palette("TextDim"),
+            HorizontalContentAlignment = HorizontalAlignment.Center,
+
+            // ◆ NOT FOCUSABLE, AND THAT IS THE WHOLE CONTROL RATHER THAN A DETAIL.
+            //
+            // These keys are an extension of the keyboard, not a place the keyboard should go. As
+            // ordinary buttons they took Avalonia's focus on tap, which takes it off the NativeWebView
+            // — and the package's own OnLostFocus then calls the adapter's ResignFocus(). So pressing
+            // Tab or an arrow handed the terminal one byte and took the keyboard away from it: the next
+            // thing typed on a hardware keyboard went nowhere, and the row went on working because its
+            // buttons are pressed rather than typed into, which is what makes it look like the terminal
+            // had died instead.
+            //
+            // Focusable=false is what a toolbar button is, and it means the focused element never
+            // changes: the WebView is still it, so nothing resigns and nothing has to be handed back.
+            //
+            // At the Avalonia layer, that is. Android keeps a focus of its own, and the touch that
+            // presses one of these keys hands it to Avalonia's input view regardless of what Avalonia
+            // decides about its element — taking the keyboard's input connection off the terminal and
+            // swapping its layout mid-typing. The row's own pointer handlers in BuildAccessoryRow hand
+            // that half back; see TerminalFocus for the whole story, including why the handing back has
+            // to be posted rather than done inline.
+            Focusable = false,
+        };
+
     private void ToggleControl()
     {
         controlLatched = !controlLatched;

src/DodoSSH.Client.App/App.axaml

@@ -153,6 +153,7 @@
       <Setter Property="CornerRadius" Value="6" />
       <Setter Property="Padding" Value="6,2" />
       <Setter Property="MinHeight" Value="0" />
+      <Setter Property="VerticalContentAlignment" Value="Center" />
       <Setter Property="Foreground" Value="{StaticResource TextDim}" />
       <Setter Property="FontFamily" Value="{StaticResource MonoFont}" />
       <Setter Property="FontWeight" Value="Medium" />
@@ -463,6 +464,7 @@
       <Setter Property="Padding" Value="13,7" />
       <Setter Property="HorizontalAlignment" Value="Stretch" />
       <Setter Property="HorizontalContentAlignment" Value="Left" />
+      <Setter Property="VerticalContentAlignment" Value="Center" />
       <Setter Property="CornerRadius" Value="12" />
     </Style>
     <Style Selector="Button.navuser /template/ ContentPresenter#PART_ContentPresenter">
@@ -482,6 +484,7 @@
     <Style Selector="Button.poprow">
       <Setter Property="HorizontalAlignment" Value="Stretch" />
       <Setter Property="HorizontalContentAlignment" Value="Stretch" />
+      <Setter Property="VerticalContentAlignment" Value="Center" />
       <Setter Property="Padding" Value="11,4" />
       <Setter Property="CornerRadius" Value="8" />
       <Setter Property="MinHeight" Value="20" />
@@ -671,6 +674,7 @@
     -->
     <Style Selector="Button.choice">
       <Setter Property="Padding" Value="10,5" />
+      <Setter Property="VerticalContentAlignment" Value="Center" />
       <Setter Property="FontFamily" Value="{StaticResource MonoFont}" />
       <Setter Property="FontSize" Value="10.5" />
       <Setter Property="LetterSpacing" Value="0.5" />
@@ -1120,6 +1124,7 @@
     <Style Selector="Button.panechip">
       <Setter Property="Padding" Value="8,4" />
       <Setter Property="MinHeight" Value="0" />
+      <Setter Property="VerticalContentAlignment" Value="Center" />
       <Setter Property="FontFamily" Value="{StaticResource MonoFont}" />
       <Setter Property="FontSize" Value="11" />
       <Setter Property="FontWeight" Value="Medium" />

src/DodoSSH.Client.App/Assets/dodossh-icon.ps1

@@ -1,13 +1,18 @@
-# Regenerates dodossh.ico from the same geometry the Android launcher icon draws.
+# Regenerates dodossh.ico and dodossh.icns from the same geometry the Android launcher icon draws.
 #
 # The phone's mark is a vector — Resources/drawable/ic_launcher_foreground.xml — and the whole
 # reason it is a vector is that there is then one geometry to change and no set of PNG densities
-# to forget one of. Windows will not take a vector: <ApplicationIcon> wants an .ico and nothing
-# else, and Window.Icon wants a bitmap. So the raster exists, and this script is how it stays
-# honest: the numbers below are the ones in that XML, and regenerating is the whole edit.
+# to forget one of. Neither desktop platform will take a vector: <ApplicationIcon> wants an .ico
+# and nothing else, Window.Icon wants a bitmap, and vpk wants an .icns for the macOS bundle. So
+# the rasters exist, and this script is how they stay honest: the numbers below are the ones in
+# that XML, and regenerating is the whole edit.
 #
 #     pwsh -File src/DodoSSH.Client.App/Assets/dodossh-icon.ps1
 #
+# Both outputs are written every run, deliberately. Two scripts, or one script with a switch,
+# is how the two files come to be drawn from different geometry — which nobody would notice,
+# because no one person looks at a Windows taskbar and a macOS Dock on the same afternoon.
+#
 # Coordinates are the launcher's 108-unit viewport, mapped so the middle 72 fills the canvas.
 # That 72 is not an arbitrary crop: it is the part of an adaptive icon a launcher actually shows,
 # the outer 18 on each edge being what it eats for masking and parallax. Rendering the whole 108
@@ -30,29 +35,49 @@ $ink = [System.Drawing.ColorTranslator]::FromHtml('#FFFFFF')     # AccentInk
 # the gap, and its downsampler is not kind to a hairline.
 $sizes = @(16, 20, 24, 32, 40, 48, 64, 128, 256)
 
-function New-MarkPng([int]$size)
+# $tileFraction is how much of the canvas the accent tile fills, and it is the one number that
+# differs between the two platforms.
+#
+# Windows passes 1.0: the tile bleeds to the edge, because Windows draws application icons at
+# whatever size they come in and every other icon on the taskbar does the same.
+#
+# macOS passes 0.8047, and that is not taste. Apple's icon grid puts a rounded-rect app icon in
+# an 824-pixel square inside a 1024-pixel canvas — 824/1024 — with the remaining hundred pixels a
+# side left as air for the Dock's shadow and its magnification. An icon that ignores the grid and
+# bleeds to the edge does not read as bold; it reads as the one icon in the Dock that is too big,
+# because it sits beside Finder and Safari which do not.
+function New-MarkPng([int]$size, [double]$tileFraction = 1.0)
 {
     $bitmap = New-Object System.Drawing.Bitmap($size, $size, [System.Drawing.Imaging.PixelFormat]::Format32bppArgb)
     $g = [System.Drawing.Graphics]::FromImage($bitmap)
     $g.SmoothingMode = [System.Drawing.Drawing2D.SmoothingMode]::AntiAlias
     $g.PixelOffsetMode = [System.Drawing.Drawing2D.PixelOffsetMode]::HighQuality
 
+    # The tile, and the inset that centres it when it does not fill the canvas.
+    $tile = [double]$size * $tileFraction
+    $inset = ([double]$size - $tile) / 2.0
+
     # The accent tile, rounded as a launcher mask rounds it. A square-cornered tile would be the
     # one icon on the taskbar with corners, which reads as unfinished rather than as deliberate.
-    $radius = [double]$size * 0.22
+    #
+    # 0.22 of the tile rather than of the canvas, so the corner keeps its proportion to the shape
+    # it is rounding instead of growing as the air around it does. It is also within a whisker of
+    # the 185/824 Apple's own grid specifies, which is why one radius serves both files.
+    $radius = $tile * 0.22
     $d = $radius * 2.0
     $path = New-Object System.Drawing.Drawing2D.GraphicsPath
-    $path.AddArc(0.0, 0.0, $d, $d, 180, 90)
-    $path.AddArc($size - $d, 0.0, $d, $d, 270, 90)
-    $path.AddArc($size - $d, $size - $d, $d, $d, 0, 90)
-    $path.AddArc(0.0, $size - $d, $d, $d, 90, 90)
+    $path.AddArc($inset, $inset, $d, $d, 180, 90)
+    $path.AddArc($inset + $tile - $d, $inset, $d, $d, 270, 90)
+    $path.AddArc($inset + $tile - $d, $inset + $tile - $d, $d, $d, 0, 90)
+    $path.AddArc($inset, $inset + $tile - $d, $d, $d, 90, 90)
     $path.CloseFigure()
     $brush = New-Object System.Drawing.SolidBrush($accent)
     $g.FillPath($brush, $path)
 
-    # 108-viewport units to pixels, with the outer 18 dropped on each edge.
-    $scale = [double]$size / 72.0
-    function P([double]$x, [double]$y) { New-Object System.Drawing.PointF((($x - 18.0) * $scale), (($y - 18.0) * $scale)) }
+    # 108-viewport units to pixels, with the outer 18 dropped on each edge. Scaled to the tile and
+    # offset by the inset, so the glyph keeps its place within the tile at either fraction.
+    $scale = $tile / 72.0
+    function P([double]$x, [double]$y) { New-Object System.Drawing.PointF((($x - 18.0) * $scale + $inset), (($y - 18.0) * $scale + $inset)) }
 
     # A stroke thinner than a pixel renders as a grey suggestion of itself, which at 16px is the
     # difference between a mark and a smudge. The phone's file already bumps this width for the
@@ -117,3 +142,86 @@ $target = Join-Path $PSScriptRoot 'dodossh.ico'
 $w.Dispose(); $out.Dispose()
 
 Write-Output "Wrote $target ($($sizes.Count) sizes, $((Get-Item $target).Length) bytes)"
+
+# ---- dodossh.icns, for the macOS bundle ----------------------------------------------------------
+#
+# Written here rather than by `iconutil` on a Mac, and that is the point of doing it the long way.
+# iconutil is the documented tool and it exists only on macOS, so an icon that needed it could not
+# be regenerated on the machine this project is developed on — the geometry above would change and
+# the .icns would quietly keep the old mark until somebody next opened a Mac. The container format
+# is a magic word, a length and a run of typed PNG chunks, which is little enough to own.
+#
+# ◆ EVERY LENGTH IN THIS FILE IS BIG-ENDIAN, AND BinaryWriter IS NOT.
+#
+# The one thing that will catch anybody editing this. A .icns written little-endian is not rejected
+# with an error — Finder and vpk both just show the placeholder icon, because the first chunk claims
+# a length of about two billion and the parser walks off the end and gives up. Hence Write-BE32.
+#
+# Type codes are Apple's, and the pairs are not redundant. ic08 and ic13 are both 256 pixels because
+# one is "256 at 1x" and the other is "128 at 2x", and a Retina display asked for the second will not
+# accept the first. Same for ic09/ic14 at 512. iconutil emits both from an .iconset for this reason,
+# so this does too.
+$icnsTypes = @(
+    @{ Type = 'ic11'; Size = 32 }    # 16@2x
+    @{ Type = 'ic12'; Size = 64 }    # 32@2x
+    @{ Type = 'ic07'; Size = 128 }   # 128@1x
+    @{ Type = 'ic13'; Size = 256 }   # 128@2x
+    @{ Type = 'ic08'; Size = 256 }   # 256@1x
+    @{ Type = 'ic14'; Size = 512 }   # 256@2x
+    @{ Type = 'ic09'; Size = 512 }   # 512@1x
+    @{ Type = 'ic10'; Size = 1024 }  # 512@2x
+)
+
+# Apple's icon grid: an 824-pixel shape centred in a 1024-pixel canvas. See New-MarkPng.
+$macTileFraction = 824.0 / 1024.0
+
+# Rendered once per distinct pixel size rather than once per type code, so the two 256s and the two
+# 512s are byte-identical and the file does not carry the same image twice over at different
+# compression. It also halves the drawing, which at 1024 is not nothing.
+$rendered = @{}
+foreach ($size in ($icnsTypes.Size | Sort-Object -Unique))
+{
+    [byte[]]$png = New-MarkPng $size $macTileFraction
+    $rendered[$size] = $png
+}
+
+$icns = New-Object System.IO.MemoryStream
+
+function Write-BE32([System.IO.Stream]$stream, [uint32]$value)
+{
+    $bytes = [System.BitConverter]::GetBytes($value)
+    if ([System.BitConverter]::IsLittleEndian) { [array]::Reverse($bytes) }
+    $stream.Write($bytes, 0, 4)
+}
+
+function Write-Ascii([System.IO.Stream]$stream, [string]$text)
+{
+    $bytes = [System.Text.Encoding]::ASCII.GetBytes($text)
+    $stream.Write($bytes, 0, $bytes.Length)
+}
+
+# The header's length field covers the whole file including the header, so it is written last —
+# eight bytes of nothing now, seeked back to and filled in once the total is known.
+Write-Ascii $icns 'icns'
+Write-BE32 $icns 0
+
+foreach ($entry in $icnsTypes)
+{
+    $payload = $rendered[$entry.Size]
+    Write-Ascii $icns $entry.Type
+
+    # Length includes this chunk's own eight-byte header, which is the off-by-eight everybody
+    # writes once.
+    Write-BE32 $icns ([uint32]($payload.Length + 8))
+    $icns.Write($payload, 0, $payload.Length)
+}
+
+$total = [uint32]$icns.Length
+$icns.Position = 4
+Write-BE32 $icns $total
+
+$icnsTarget = Join-Path $PSScriptRoot 'dodossh.icns'
+[System.IO.File]::WriteAllBytes($icnsTarget, $icns.ToArray())
+$icns.Dispose()
+
+Write-Output "Wrote $icnsTarget ($($icnsTypes.Count) entries, $((Get-Item $icnsTarget).Length) bytes)"

src/DodoSSH.Client.App/DodoSSH.Client.App.csproj

@@ -66,6 +66,21 @@
     -->
     <DodoChannel Condition="'$(DodoChannel)' == ''">release</DodoChannel>
 
+    <!--
+      For the macOS keychain interop in Platform/, and for nothing else.
+
+      Set on this project rather than in Directory.Build.props deliberately. The frameworks that hold a
+      Secure Enclave key take CFDictionaries of raw pointers, so building one means pinning arrays and
+      taking their addresses — see MacDeviceKeyStore. Every other project here is managed code with no
+      business doing that, and a solution-wide flag would quietly permit it everywhere, including in the
+      crypto project where a stray pointer is the last thing anybody wants to have been allowed.
+
+      The alternative — GCHandle.Alloc with GCHandleType.Pinned — needs no flag and was considered. It
+      would replace each `fixed` with an allocate/free pair that has to be balanced by hand across the
+      early returns those methods are full of, which trades a compiler-checked scope for a manual one.
+    -->
+    <AllowUnsafeBlocks>true</AllowUnsafeBlocks>
+
     <!--
       False here, unlike every server project. The root Directory.Build.props sets it true because
       the API is container-hosted, UTC-only and has no business formatting anything for a human.

src/DodoSSH.Client.App/Platform/MacDeviceKeyStore.cs

@@ -0,0 +1,598 @@
+using System.Runtime.InteropServices;
+using System.Runtime.Versioning;
+using System.Text;
+using DodoSSH.Client.Session;
+using static DodoSSH.Client.App.Platform.MacSecurity;
+
+namespace DodoSSH.Client.App.Platform;
+
+/// <summary>
+/// Keeps the device key encrypted to a Secure Enclave key whose use requires the user's presence.
+/// </summary>
+/// <remarks>
+/// <para>
+/// The macOS counterpart of <see cref="WindowsDeviceKeyStore"/>, and the same argument holds it up:
+/// <b>the consent is enforced by the platform, not by this class</b>. The unwrapping key is generated
+/// inside the Secure Enclave and never leaves it — there is no code path, privileged or otherwise, that
+/// turns it into bytes — and it is created under an access control requiring
+/// <see cref="AccessControlFlags.UserPresence"/>, so Touch ID or the login password is a condition of
+/// <em>using</em> it. Malware running as the user can ask for a decryption; it cannot answer the prompt,
+/// and the attempt is visible.
+/// </para>
+/// <para>
+/// A store that showed its own prompt and then read a protected file would be trivially bypassed, which
+/// is the mistake ADR 0007 originally described and the Windows store's comment corrects. The correction
+/// applies here unchanged.
+/// </para>
+/// <para>
+/// <b>P-256 and ECIES, where Windows uses RSA-OAEP, and the difference is not a preference.</b> The
+/// Secure Enclave holds exactly one kind of key: a 256-bit key on the NIST P-256 curve. It will not hold
+/// an RSA key at any size. So the wrap is <c>eciesEncryptionCofactorX963SHA256AESGCM</c> — an ephemeral
+/// agreement against the enclave's public half, X9.63-KDF to an AES-GCM key, and the ephemeral public
+/// key carried in the output. The framework does all of that; what matters here is that the input is 32
+/// bytes and there is no size limit worth worrying about.
+/// </para>
+/// <para>
+/// <b>Sealing is silent and unsealing prompts, which is better than the Windows shape rather than merely
+/// different.</b> On Windows, <c>CngKey.Create</c> with <c>ProtectKey</c> raises a dialog at creation as
+/// well, because the policy means "protect this key with a PIN" and Windows sets that up there and then.
+/// Here <see cref="SecKeyCopyPublicKey"/> works on an enclave key without any prompt, so registering a
+/// device shows nothing and only unlock asks. <see cref="SaveAsync"/> is therefore not user-facing on
+/// this platform — but it is still called from where the Windows one has to be, and relying on that
+/// difference would make the shared caller platform-specific for no gain.
+/// </para>
+/// <para>
+/// <b>What this cannot be tested against, and what follows from that.</b> Every method except
+/// <see cref="IsAvailableAsync"/> and the empty case of <see cref="TryLoadAsync"/> needs an interactive
+/// login session and real enclave hardware, so none can be exercised by an automated test — the same
+/// line the Windows store draws. It also means <see cref="IsSupported"/> must probe rather than infer:
+/// see its remarks for the three ordinary machines that have no usable enclave and must degrade to the
+/// passphrase rather than fail at unlock.
+/// </para>
+/// </remarks>
+[SupportedOSPlatform("macos")]
+public sealed partial class MacDeviceKeyStore : IDeviceKeyStore
+{
+    /// <summary>
+    /// The keychain tag this application's enclave key is filed under.
+    /// </summary>
+    /// <remarks>
+    /// Versioned for the reason the Windows key name is: a future change of curve or wrap algorithm can
+    /// create a new key beside the old one rather than failing to open blobs written by a previous
+    /// build. A device that cannot be opened falls back to the passphrase, which is survivable — but
+    /// silently, and a user would only notice their fingerprint had stopped working.
+    ///
+    /// Prefixed with the bundle identifier because the keychain is shared across every application the
+    /// user runs, unlike a CNG key name, which is scoped to the user's key store already.
+    /// </remarks>
+    private const string KeyTag = "dev.dodotech.dodossh.devicekey.v1";
+
+    /// <summary>
+    /// Shown in the Touch ID prompt, so it has to read as a sentence to a person.
+    /// </summary>
+    /// <remarks>
+    /// macOS composes it into "DodoSSH is trying to ...", so this is a verb phrase and not a sentence of
+    /// its own. The same words the Windows consent dialog uses.
+    /// </remarks>
+    private const string ConsentPrompt = "unlock your DodoSSH vault";
+
+    private readonly ClientPaths paths;
+
+    /// <summary>Creates the store.</summary>
+    public MacDeviceKeyStore(ClientPaths paths)
+    {
+        ArgumentNullException.ThrowIfNull(paths);
+        this.paths = paths;
+    }
+
+    /// <summary>
+    /// Whether this Mac has a Secure Enclave that will hold a key for this build.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// Probed by creating a throwaway key and deleting it, rather than by asking whether the hardware
+    /// exists. Three ordinary situations answer "no" here and would otherwise only be discovered at the
+    /// moment somebody tried to unlock:
+    /// </para>
+    /// <para>
+    /// <b>An Intel Mac with no T2.</b> Apple Silicon and T2 machines have an enclave; earlier Intel
+    /// models do not, and there is no single attribute that says so.
+    /// </para>
+    /// <para>
+    /// <b>A build that is not code signed.</b> Enclave key creation requires a signing identity, so
+    /// every <c>dotnet run</c> and every build from an IDE fails here with a missing-entitlement error.
+    /// That is the correct answer rather than a nuisance: a development build should keep asking for the
+    /// passphrase, and this is what makes it do so without a platform check somewhere else.
+    /// </para>
+    /// <para>
+    /// <b>A machine with no login password set.</b> <see cref="AccessControlFlags.UserPresence"/> has
+    /// nothing to demand, and the framework refuses the access control object rather than silently
+    /// creating a key anybody could use.
+    /// </para>
+    /// <para>
+    /// The probe uses its own tag and no UI policy, so nothing prompts and nothing collides with the
+    /// real key. It is deleted immediately; a probe key left behind would accumulate one per launch.
+    /// </para>
+    /// </remarks>
+    internal static bool IsSupported()
+    {
+        try
+        {
+            var probe = $"{KeyTag}.probe.{Guid.CreateVersion7():N}";
+
+            using var scope = new CoreFoundationScope();
+
+            var symbols = MacSymbols.Resolve();
+
+            if (!symbols.Complete)
+            {
+                return false;
+            }
+
+            var key = CreateEnclaveKey(scope, symbols, probe);
+
+            if (key == IntPtr.Zero)
+            {
+                return false;
+            }
+
+            // Discarded deliberately. The question this method answers is whether the enclave will make a
+            // key, and it demonstrably just did; a failure to clean the probe up afterwards leaves one
+            // stray keychain item and does not make the answer no.
+            _ = DeleteKey(symbols, probe);
+
+            return true;
+        }
+        catch (Exception exception) when (exception is DllNotFoundException
+                                              or EntryPointNotFoundException
+                                              or BadImageFormatException)
+        {
+            // A macOS without these frameworks is not a thing that exists, so this is really the guard
+            // for the case that does: a future release renaming or removing one of them. The answer is
+            // the same as for hardware that is absent — no device key, ask for the passphrase.
+            return false;
+        }
+    }
+
+    /// <inheritdoc />
+    public ValueTask<bool> IsAvailableAsync(CancellationToken cancellationToken) =>
+        ValueTask.FromResult(IsSupported());
+
+    /// <inheritdoc />
+    public async ValueTask SaveAsync(
+        ReadOnlyMemory<byte> devicePrivateKey,
+        CancellationToken cancellationToken)
+    {
+        var sealedKey = Seal(devicePrivateKey.Span)
+            ?? throw new InvalidOperationException(
+                "The Secure Enclave would not seal the device key. Check IsAvailableAsync before offering to register one.");
+
+        paths.EnsureCreated();
+
+        await File.WriteAllBytesAsync(paths.DeviceKeyFile, sealedKey, cancellationToken)
+            .ConfigureAwait(false);
+    }
+
+    /// <inheritdoc />
+    public async ValueTask<byte[]?> TryLoadAsync(CancellationToken cancellationToken)
+    {
+        if (!File.Exists(paths.DeviceKeyFile))
+        {
+            return null;
+        }
+
+        var sealedKey = await File.ReadAllBytesAsync(paths.DeviceKeyFile, cancellationToken)
+            .ConfigureAwait(false);
+
+        return Unseal(sealedKey);
+    }
+
+    /// <inheritdoc />
+    public ValueTask ForgetAsync(CancellationToken cancellationToken)
+    {
+        if (File.Exists(paths.DeviceKeyFile))
+        {
+            File.Delete(paths.DeviceKeyFile);
+        }
+
+        var symbols = MacSymbols.Resolve();
+
+        if (symbols.Complete)
+        {
+            // Discarded, and that is deliberate: there is nothing a caller could do about a failure here,
+            // and the file deleted above is the half that decides whether unlock will try at all. A key
+            // left in the enclave with no ciphertext to open is inert.
+            _ = DeleteKey(symbols, KeyTag);
+        }
+
+        return ValueTask.CompletedTask;
+    }
+
+    /// <remarks>
+    /// Silent: it uses only the public half. Null on every failure, and the caller's answer to all of
+    /// them is the same — do not offer a device unlock.
+    /// </remarks>
+    private static byte[]? Seal(ReadOnlySpan<byte> devicePrivateKey)
+    {
+        try
+        {
+            using var scope = new CoreFoundationScope();
+
+            var symbols = MacSymbols.Resolve();
+
+            if (!symbols.Complete)
+            {
+                return null;
+            }
+
+            // Created on first use rather than at registration, so that a device key re-registered after
+            // a ForgetAsync gets a key again without anything having to notice that it had gone.
+            var privateKey = FindKey(scope, symbols, KeyTag, prompt: null);
+
+            if (privateKey == IntPtr.Zero)
+            {
+                privateKey = CreateEnclaveKey(scope, symbols, KeyTag);
+            }
+
+            if (privateKey == IntPtr.Zero)
+            {
+                return null;
+            }
+
+            var publicKey = scope.Keep(SecKeyCopyPublicKey(privateKey));
+
+            if (publicKey == IntPtr.Zero)
+            {
+                return null;
+            }
+
+            var plaintext = Data(scope, devicePrivateKey);
+
+            if (plaintext == IntPtr.Zero)
+            {
+                return null;
+            }
+
+            var ciphertext = scope.Keep(
+                SecKeyCreateEncryptedData(publicKey, symbols.EciesAlgorithm, plaintext, out var error));
+
+            scope.Keep(error);
+
+            return ciphertext == IntPtr.Zero ? null : ToArray(ciphertext);
+        }
+        catch (Exception exception) when (exception is DllNotFoundException
+                                              or EntryPointNotFoundException
+                                              or BadImageFormatException)
+        {
+            return null;
+        }
+    }
+
+    /// <remarks>
+    /// <para>
+    /// This is the call that prompts. Every failure becomes null, and the set is wider than it looks:
+    /// the key may be gone, the user may have cancelled or let the prompt time out, the enclave may have
+    /// invalidated it after the login password was reset, or the blob may predate a key that has since
+    /// been replaced. None of them are distinguishable to a user and all have the same remedy, so none
+    /// are worth telling apart here — see <c>UnlockStatus.DeviceKeyUnavailable</c>.
+    /// </para>
+    /// <para>
+    /// Blocking, and it blocks on a person. The prompt is modal to the application, so this must not run
+    /// on a thread that is also expected to draw the window behind it.
+    /// </para>
+    /// </remarks>
+    private static byte[]? Unseal(byte[] sealedKey)
+    {
+        try
+        {
+            using var scope = new CoreFoundationScope();
+
+            var symbols = MacSymbols.Resolve();
+
+            if (!symbols.Complete)
+            {
+                return null;
+            }
+
+            var privateKey = FindKey(scope, symbols, KeyTag, ConsentPrompt);
+
+            if (privateKey == IntPtr.Zero)
+            {
+                return null;
+            }
+
+            var ciphertext = Data(scope, sealedKey);
+
+            if (ciphertext == IntPtr.Zero)
+            {
+                return null;
+            }
+
+            var plaintext = scope.Keep(
+                SecKeyCreateDecryptedData(privateKey, symbols.EciesAlgorithm, ciphertext, out var error));
+
+            scope.Keep(error);
+
+            return plaintext == IntPtr.Zero ? null : ToArray(plaintext);
+        }
+        catch (Exception exception) when (exception is DllNotFoundException
+                                              or EntryPointNotFoundException
+                                              or BadImageFormatException)
+        {
+            return null;
+        }
+    }
+
+    /// <summary>
+    /// Generates a key inside the Secure Enclave, filed under <paramref name="tag"/>. Owned by the scope.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// The attribute dictionary is the whole security decision, so it is worth reading rather than
+    /// pattern-matching. <c>TokenID = SecureEnclave</c> is what puts the private half in hardware;
+    /// without it this silently generates an ordinary software key that behaves identically in every
+    /// visible way and protects nothing.
+    /// </para>
+    /// <para>
+    /// <c>AccessibleWhenUnlockedThisDeviceOnly</c> rather than any of the migratable classes, because a
+    /// device key that could be restored onto another machine from a backup would no longer mean "this
+    /// machine". The enclave already makes that impossible; saying it as well means the intent survives
+    /// a future change of storage.
+    /// </para>
+    /// <para>
+    /// <c>UseDataProtectionKeychain</c> is the macOS-specific one and the easiest to omit. Without it,
+    /// macOS routes this to the older file-based keychain, which does not understand access control
+    /// objects or the enclave, and the call fails with a parameter error that says nothing about the
+    /// missing key.
+    /// </para>
+    /// </remarks>
+    private static IntPtr CreateEnclaveKey(CoreFoundationScope scope, MacSymbols symbols, string tag)
+    {
+        var access = scope.Keep(SecAccessControlCreateWithFlags(
+            IntPtr.Zero,
+            symbols.AccessibleWhenUnlockedThisDeviceOnly,
+            AccessControlFlags.PrivateKeyUsage | AccessControlFlags.UserPresence,
+            out var accessError));
+
+        scope.Keep(accessError);
+
+        if (access == IntPtr.Zero)
+        {
+            return IntPtr.Zero;
+        }
+
+        var privateAttrs = Dictionary(
+            scope,
+            [symbols.AttrIsPermanent, symbols.AttrApplicationTag, symbols.AttrAccessControl],
+            [symbols.True, TagData(scope, tag), access]);
+
+        if (privateAttrs == IntPtr.Zero)
+        {
+            return IntPtr.Zero;
+        }
+
+        var keySize = Number(scope, 256);
+
+        var parameters = Dictionary(
+            scope,
+            [
+                symbols.AttrKeyType,
+                symbols.AttrKeySizeInBits,
+                symbols.AttrTokenId,
+                symbols.UseDataProtectionKeychain,
+                symbols.PrivateKeyAttrs,
+            ],
+            [
+                symbols.KeyTypeEcSecPrimeRandom,
+                keySize,
+                symbols.TokenIdSecureEnclave,
+                symbols.True,
+                privateAttrs,
+            ]);
+
+        if (parameters == IntPtr.Zero)
+        {
+            return IntPtr.Zero;
+        }
+
+        var key = scope.Keep(SecKeyCreateRandomKey(parameters, out var error));
+
+        scope.Keep(error);
+
+        return key;
+    }
+
+    /// <summary>
+    /// Looks the enclave key up by tag. Owned by the scope; zero when there is none.
+    /// </summary>
+    /// <remarks>
+    /// <paramref name="prompt"/> is attached here and consumed later: the lookup itself does not raise
+    /// anything, because a handle to an enclave key is not a use of it. The words reach the user at the
+    /// decrypt, which is the operation the access control actually guards.
+    ///
+    /// <c>UseOperationPrompt</c> is deprecated in favour of an <c>LAContext</c>, and is used anyway. An
+    /// LAContext would mean binding LocalAuthentication as well for one string, and the deprecated key
+    /// still works; the day it stops, this call fails and the store degrades to the passphrase, which is
+    /// the failure this whole class is built to degrade into.
+    /// </remarks>
+    private static IntPtr FindKey(CoreFoundationScope scope, MacSymbols symbols, string tag, string? prompt)
+    {
+        List<IntPtr> keys =
+        [
+            symbols.Class,
+            symbols.AttrApplicationTag,
+            symbols.AttrKeyType,
+            symbols.UseDataProtectionKeychain,
+            symbols.ReturnRef,
+        ];
+
+        List<IntPtr> values =
+        [
+            symbols.ClassKey,
+            TagData(scope, tag),
+            symbols.KeyTypeEcSecPrimeRandom,
+            symbols.True,
+            symbols.True,
+        ];
+
+        if (prompt is not null)
+        {
+            keys.Add(symbols.UseOperationPrompt);
+            values.Add(scope.Keep(CFString(prompt)));
+        }
+
+        var query = Dictionary(scope, [.. keys], [.. values]);
+
+        if (query == IntPtr.Zero)
+        {
+            return IntPtr.Zero;
+        }
+
+        var status = SecItemCopyMatching(query, out var result);
+
+        // errSecItemNotFound is the ordinary answer on a machine that has never registered a device, and
+        // it is not distinguished from any other failure for the reason the class remarks give.
+        return status == Success ? scope.Keep(result) : IntPtr.Zero;
+    }
+
+    /// <summary>Removes the key with this tag from the keychain.</summary>
+    /// <returns>Whether the keychain now has no key under this tag.</returns>
+    /// <remarks>
+    /// <c>ItemNotFound</c> counts as success, and that is the common case rather than an edge: it is
+    /// what a machine that never registered a device answers, and what the second of two
+    /// <see cref="ForgetAsync"/> calls answers. Treating it as a failure would make forgetting a device
+    /// twice report a problem that does not exist.
+    /// </remarks>
+    private static bool DeleteKey(MacSymbols symbols, string tag)
+    {
+        using var scope = new CoreFoundationScope();
+
+        var query = Dictionary(
+            scope,
+            [symbols.Class, symbols.AttrApplicationTag, symbols.UseDataProtectionKeychain],
+            [symbols.ClassKey, TagData(scope, tag), symbols.True]);
+
+        if (query == IntPtr.Zero)
+        {
+            return false;
+        }
+
+        var status = SecItemDelete(query);
+
+        return status is Success or ItemNotFound;
+    }
+
+    // ---- Small CoreFoundation conveniences ---------------------------------------------------------
+
+    /// <remarks>
+    /// The arrays are pinned for the duration of the call and not beyond it, which is correct because
+    /// <c>CFDictionaryCreate</c> copies them: the dictionary retains each key and value, and never reads
+    /// the arrays again.
+    /// </remarks>
+    private static IntPtr Dictionary(CoreFoundationScope scope, IntPtr[] keys, IntPtr[] values)
+    {
+        // A zero anywhere means one of the constants did not resolve or an earlier allocation failed.
+        // Passing it on produces a dictionary with a null key, which CFDictionaryCreate does not reject
+        // — it crashes inside the callback table instead.
+        if (Array.IndexOf(keys, IntPtr.Zero) >= 0 || Array.IndexOf(values, IntPtr.Zero) >= 0)
+        {
+            return IntPtr.Zero;
+        }
+
+        var symbols = MacSymbols.Resolve();
+
+        unsafe
+        {
+            fixed (IntPtr* keyPtr = keys)
+            fixed (IntPtr* valuePtr = values)
+            {
+                return scope.Keep(CFDictionaryCreate(
+                    IntPtr.Zero,
+                    (IntPtr)keyPtr,
+                    (IntPtr)valuePtr,
+                    keys.Length,
+                    symbols.TypeDictionaryKeyCallBacks,
+                    symbols.TypeDictionaryValueCallBacks));
+            }
+        }
+    }
+
+    /// <summary>Copies bytes into a CFData. Owned by the scope.</summary>
+    /// <remarks>
+    /// The pin lasts only as long as the call, which is correct: <c>CFDataCreate</c> copies, so the
+    /// CFData does not reference this memory afterwards. <c>CFDataCreateWithBytesNoCopy</c> would not,
+    /// and is not used for exactly that reason — it would hand the framework a pointer into the managed
+    /// heap and rely on the object staying where the collector first put it.
+    /// </remarks>
+    private static IntPtr Data(CoreFoundationScope scope, ReadOnlySpan<byte> bytes)
+    {
+        unsafe
+        {
+            fixed (byte* pointer = bytes)
+            {
+                return scope.Keep(CFDataCreate(IntPtr.Zero, (IntPtr)pointer, bytes.Length));
+            }
+        }
+    }
+
+    /// <remarks>
+    /// UTF-8 rather than any other encoding, and it only has to be consistent with itself: the tag is an
+    /// opaque blob the keychain matches byte for byte, so what matters is that a lookup encodes it the
+    /// same way the creation did. It is written once, here, for exactly that reason.
+    /// </remarks>
+    private static IntPtr TagData(CoreFoundationScope scope, string tag) =>
+        Data(scope, Encoding.UTF8.GetBytes(tag));
+
+    private static IntPtr Number(CoreFoundationScope scope, int value)
+    {
+        unsafe
+        {
+            return scope.Keep(CFNumberCreate(IntPtr.Zero, (nint)CFNumberIntType, (IntPtr)(&value)));
+        }
+    }
+
+    /// <summary>Builds a CFString from a managed string. Owned, so the caller tracks it.</summary>
+    /// <remarks>
+    /// Built explicitly rather than left to the marshaller, because these calls take a
+    /// <c>CFStringRef</c> and not a C string — the runtime's default marshalling would hand over a
+    /// <c>char*</c>, which CoreFoundation reads as an object pointer and follows into nothing.
+    /// </remarks>
+    private static IntPtr CFString(string value)
+    {
+        var bytes = Encoding.UTF8.GetBytes(value);
+
+        unsafe
+        {
+            fixed (byte* pointer = bytes)
+            {
+                // kCFStringEncodingUTF8 is 0x08000100, spelled out rather than named because it is the
+                // only encoding constant this file uses.
+                return CFStringCreateWithBytes(IntPtr.Zero, (IntPtr)pointer, bytes.Length, 0x08000100, false);
+            }
+        }
+    }
+
+    [LibraryImport(CoreFoundation)]
+    private static partial IntPtr CFStringCreateWithBytes(
+        IntPtr allocator,
+        IntPtr bytes,
+        nint numBytes,
+        uint encoding,
+        [MarshalAs(UnmanagedType.U1)] bool isExternalRepresentation);
+
+    private static byte[] ToArray(IntPtr data)
+    {
+        var length = (int)CFDataGetLength(data);
+        var pointer = CFDataGetBytePtr(data);
+
+        if (length <= 0 || pointer == IntPtr.Zero)
+        {
+            return [];
+        }
+
+        var result = new byte[length];
+        Marshal.Copy(pointer, result, 0, length);
+
+        return result;
+    }
+}

src/DodoSSH.Client.App/Platform/MacSecurity.cs

@@ -0,0 +1,254 @@
+using System.Runtime.InteropServices;
+using System.Runtime.Versioning;
+
+namespace DodoSSH.Client.App.Platform;
+
+/// <summary>
+/// The pieces of CoreFoundation and Security.framework <see cref="MacDeviceKeyStore"/> needs.
+/// </summary>
+/// <remarks>
+/// <para>
+/// Separated from the store itself because it is a different kind of code with a different kind of
+/// review: nothing here makes a decision, and everything here is a translation of a C declaration that
+/// is either right or wrong. Mixing the two would mean the security argument in
+/// <see cref="MacDeviceKeyStore"/> had to be read past two hundred lines of marshalling to find.
+/// </para>
+/// <para>
+/// <b>Every Create or Copy returns an object this process owns.</b> That is CoreFoundation's Create
+/// Rule, and it is the thing here that goes wrong silently: the enclave key handle is small, so a leak
+/// shows up as nothing at all until a long-running process has done a few thousand unlocks.
+/// <see cref="CoreFoundationScope"/> exists so ownership is tracked by construction rather than by
+/// remembering, and every function below that returns a handle says whether it is owned.
+/// </para>
+/// <para>
+/// <b>The integer widths are the part worth checking against the headers rather than skimming.</b>
+/// <c>CFIndex</c>, <c>CFOptionFlags</c> and <c>CFNumberType</c> are all pointer-width on a 64-bit Mac,
+/// not 32-bit, and getting one wrong does not fail cleanly — it shifts every argument after it, so the
+/// call receives plausible rubbish and returns a parameter error that names nothing.
+/// </para>
+/// </remarks>
+[SupportedOSPlatform("macos")]
+internal static partial class MacSecurity
+{
+    internal const string SecurityFramework =
+        "/System/Library/Frameworks/Security.framework/Security";
+
+    internal const string CoreFoundation =
+        "/System/Library/Frameworks/CoreFoundation.framework/CoreFoundation";
+
+    /// <summary>
+    /// The access control flags <c>SecAccessControlCreateWithFlags</c> takes.
+    /// </summary>
+    /// <remarks>
+    /// <c>ulong</c> because the parameter is a <c>CFOptionFlags</c>, which is an <c>unsigned long</c>.
+    /// Only the two flags that are used are listed; the full set is large, and copying it in would
+    /// invite somebody to reach for one without reading what it does to the prompt — <c>Biometry</c>
+    /// alone, for instance, leaves a Mac with no Touch ID unable to unlock at all rather than falling
+    /// back to the login password.
+    /// </remarks>
+    [Flags]
+    internal enum AccessControlFlags : ulong
+    {
+        /// <summary>
+        /// Touch ID if the machine has it, the login password if not.
+        /// </summary>
+        /// <remarks>
+        /// The forgiving one, deliberately. <c>BiometryCurrentSet</c> would additionally invalidate the
+        /// key whenever a fingerprint is added or removed, which sounds stricter and here buys nothing:
+        /// this key wraps a device key whose loss already means "ask for the passphrase", so the only
+        /// effect would be users being sent back to their passphrase by an unrelated Settings change
+        /// they would never connect to it.
+        /// </remarks>
+        UserPresence = 1ul << 0,
+
+        /// <summary>Required for any key that lives in the Secure Enclave.</summary>
+        PrivateKeyUsage = 1ul << 30,
+    }
+
+    /// <summary>The CFNumberType code for a 32-bit int, from CFNumber.h.</summary>
+    internal const long CFNumberIntType = 9;
+
+    /// <summary>errSecSuccess.</summary>
+    internal const int Success = 0;
+
+    /// <summary>errSecItemNotFound, which is an answer rather than a failure.</summary>
+    internal const int ItemNotFound = -25300;
+
+    // ---- CoreFoundation ------------------------------------------------------------------------------
+
+    /// <summary>Releases an owned handle.</summary>
+    [LibraryImport(CoreFoundation)]
+    internal static partial void CFRelease(IntPtr handle);
+
+    /// <summary>Copies bytes into a new CFData. Owned.</summary>
+    [LibraryImport(CoreFoundation)]
+    internal static partial IntPtr CFDataCreate(IntPtr allocator, IntPtr bytes, nint length);
+
+    [LibraryImport(CoreFoundation)]
+    internal static partial IntPtr CFDataGetBytePtr(IntPtr data);
+
+    [LibraryImport(CoreFoundation)]
+    internal static partial nint CFDataGetLength(IntPtr data);
+
+    /// <summary>Boxes a value as a CFNumber. Owned.</summary>
+    [LibraryImport(CoreFoundation)]
+    internal static partial IntPtr CFNumberCreate(IntPtr allocator, nint theType, IntPtr valuePtr);
+
+    /// <summary>Builds an immutable dictionary. Owned.</summary>
+    /// <remarks>
+    /// <para>
+    /// The key and value arrays are passed as raw pointers to memory the caller pins, rather than as
+    /// managed arrays. Source-generated interop wants an explicit element count for a marshalled array,
+    /// and supplying one here would mean stating the length twice — once for the marshaller and once as
+    /// <paramref name="numValues"/> — which is exactly the pair that drifts.
+    /// </para>
+    /// <para>
+    /// The two callback tables are what make the dictionary retain its keys and values, which is why
+    /// they are passed rather than left null: with null callbacks the dictionary stores raw pointers and
+    /// keeps nothing alive, and the resulting use-after-free is intermittent by nature.
+    /// </para>
+    /// </remarks>
+    [LibraryImport(CoreFoundation)]
+    internal static partial IntPtr CFDictionaryCreate(
+        IntPtr allocator,
+        IntPtr keys,
+        IntPtr values,
+        nint numValues,
+        IntPtr keyCallBacks,
+        IntPtr valueCallBacks);
+
+    // ---- Security.framework --------------------------------------------------------------------------
+
+    /// <summary>Builds the access policy a Secure Enclave key is created under. Owned.</summary>
+    [LibraryImport(SecurityFramework)]
+    internal static partial IntPtr SecAccessControlCreateWithFlags(
+        IntPtr allocator,
+        IntPtr protection,
+        AccessControlFlags flags,
+        out IntPtr error);
+
+    /// <summary>Creates a key pair from an attribute dictionary. Owned.</summary>
+    [LibraryImport(SecurityFramework)]
+    internal static partial IntPtr SecKeyCreateRandomKey(IntPtr parameters, out IntPtr error);
+
+    /// <summary>The public half of a key. Owned.</summary>
+    /// <remarks>
+    /// Available even for an enclave key, and that asymmetry is the whole reason this design works: the
+    /// public half is an ordinary key this process can hold and use, while the private half is a handle
+    /// to something inside the enclave that never becomes bytes. So sealing is silent and unsealing is
+    /// the thing the user is asked about.
+    /// </remarks>
+    [LibraryImport(SecurityFramework)]
+    internal static partial IntPtr SecKeyCopyPublicKey(IntPtr key);
+
+    /// <summary>Encrypts with a public key. Owned.</summary>
+    [LibraryImport(SecurityFramework)]
+    internal static partial IntPtr SecKeyCreateEncryptedData(
+        IntPtr key,
+        IntPtr algorithm,
+        IntPtr plaintext,
+        out IntPtr error);
+
+    /// <summary>Decrypts with a private key, prompting for whatever guards it. Owned.</summary>
+    [LibraryImport(SecurityFramework)]
+    internal static partial IntPtr SecKeyCreateDecryptedData(
+        IntPtr key,
+        IntPtr algorithm,
+        IntPtr ciphertext,
+        out IntPtr error);
+
+    /// <summary>Finds a keychain item. The out handle is owned when the result is <see cref="Success"/>.</summary>
+    [LibraryImport(SecurityFramework)]
+    internal static partial int SecItemCopyMatching(IntPtr query, out IntPtr result);
+
+    /// <summary>Deletes every keychain item matching the query.</summary>
+    [LibraryImport(SecurityFramework)]
+    internal static partial int SecItemDelete(IntPtr query);
+
+    // ---- The framework constants ---------------------------------------------------------------------
+
+    /// <summary>
+    /// Reads one of a framework's global CFString constants, or zero if it is not exported.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// The keys these dictionaries take are not strings this code may spell for itself. They are
+    /// pointer-comparable constants exported by the framework, and a CFString built here with the same
+    /// characters is a different object — the lookups would miss and the call would fail with a
+    /// parameter error naming nothing.
+    /// </para>
+    /// <para>
+    /// <b>Dereferenced once, because the exported symbol is the variable rather than its value.</b>
+    /// <c>TryGetExport</c> answers the address of the global; the CFStringRef is what that address
+    /// holds. Missing the indirection produces a pointer that is stable, plausible and wrong, which is
+    /// the worst of the three available outcomes.
+    /// </para>
+    /// <para>
+    /// Zero on a missing symbol rather than an exception, because the caller's answer to every failure
+    /// is the same one — report the store unavailable and let unlock ask for the passphrase — and a
+    /// constant that has been renamed by a future macOS should reach that answer rather than a crash.
+    /// </para>
+    /// </remarks>
+    internal static IntPtr Constant(IntPtr library, string symbol) =>
+        NativeLibrary.TryGetExport(library, symbol, out var address)
+            ? Marshal.ReadIntPtr(address)
+            : IntPtr.Zero;
+}
+
+/// <summary>
+/// Releases every CoreFoundation handle put into it, in reverse order, exactly once.
+/// </summary>
+/// <remarks>
+/// <para>
+/// The alternative is a try/finally per handle, and the operations here need six or seven at a time — a
+/// dictionary holding a nested dictionary holding an access control object holding a CFData tag. Finallys
+/// nested that deep stop being read, and a handle released twice is a crash rather than a leak.
+/// </para>
+/// <para>
+/// <see cref="Keep"/> returns what it was given, so a handle can be tracked in the same expression that
+/// produces it and the call sites read as ordinary code.
+/// </para>
+/// </remarks>
+[SupportedOSPlatform("macos")]
+internal sealed class CoreFoundationScope : IDisposable
+{
+    private readonly List<IntPtr> owned = [];
+
+    private bool disposed;
+
+    /// <summary>Takes ownership of a handle and hands it straight back.</summary>
+    /// <remarks>
+    /// Zero is ignored rather than rejected. Every CoreFoundation call here answers zero on failure, so
+    /// accepting it lets a caller track the result in the expression that produces it and check it on
+    /// the next line, instead of writing the check twice.
+    /// </remarks>
+    internal IntPtr Keep(IntPtr handle)
+    {
+        if (handle != IntPtr.Zero)
+        {
+            owned.Add(handle);
+        }
+
+        return handle;
+    }
+
+    public void Dispose()
+    {
+        if (disposed)
+        {
+            return;
+        }
+
+        disposed = true;
+
+        // Reverse order, so a container is released before the things it retains. CoreFoundation does not
+        // require it — retain counts make the order irrelevant — but it keeps the lifetimes readable in a
+        // debugger, where a released container that still lists its contents is a confusing thing to meet.
+        for (var i = owned.Count - 1; i >= 0; i--)
+        {
+            MacSecurity.CFRelease(owned[i]);
+        }
+
+        owned.Clear();
+    }
+}

src/DodoSSH.Client.App/Platform/MacSymbols.cs

@@ -0,0 +1,185 @@
+using System.Runtime.InteropServices;
+using System.Runtime.Versioning;
+
+namespace DodoSSH.Client.App.Platform;
+
+/// <summary>
+/// The framework constants <see cref="MacDeviceKeyStore"/> passes to CoreFoundation and Security.
+/// </summary>
+/// <remarks>
+/// <para>
+/// Every field here is a pointer read out of a loaded framework rather than a value this code could
+/// write down. The dictionaries these go into are matched by pointer identity, so a CFString built with
+/// the same characters is a different key and the lookup misses — see <see cref="MacSecurity.Constant"/>
+/// for the indirection that trips people.
+/// </para>
+/// <para>
+/// <b>Resolved once and cached, and the caching is what makes the failure survivable.</b> Two frameworks
+/// and nineteen symbols is a lot of things to be wrong about, and the useful property is that being
+/// wrong about any one of them shows up here — as <see cref="Complete"/> being false — rather than
+/// three calls later as a parameter error. A store that reports itself unavailable sends the user back
+/// to their passphrase; a store that half works corrupts the moment somebody registers a device.
+/// </para>
+/// <para>
+/// <see cref="Lazy{T}"/> rather than a static constructor, because a type initialiser that throws
+/// poisons the type for the life of the process and turns a missing symbol into a
+/// <c>TypeInitializationException</c> at every later call site. The load is done inside a try instead,
+/// and its failure is a value.
+/// </para>
+/// </remarks>
+[SupportedOSPlatform("macos")]
+internal sealed class MacSymbols
+{
+    private static readonly Lazy<MacSymbols> Cached = new(Load, LazyThreadSafetyMode.ExecutionAndPublication);
+
+    private MacSymbols()
+    {
+    }
+
+    /// <summary>Whether every symbol resolved.</summary>
+    /// <remarks>
+    /// Checked by every caller before any of the pointers are used. It is one check rather than
+    /// nineteen, which is the only reason the call sites in <see cref="MacDeviceKeyStore"/> are
+    /// readable.
+    ///
+    /// Computed rather than stored, so that the instance returned when a framework will not load at all
+    /// — every field left at zero — answers false without that having to be set anywhere. One rule,
+    /// applied to the only state there is.
+    /// </remarks>
+    internal bool Complete => AllResolved();
+
+    // CoreFoundation.
+    internal IntPtr True { get; private init; }
+
+    internal IntPtr TypeDictionaryKeyCallBacks { get; private init; }
+
+    internal IntPtr TypeDictionaryValueCallBacks { get; private init; }
+
+    // Security: item classes and query keys.
+    internal IntPtr Class { get; private init; }
+
+    internal IntPtr ClassKey { get; private init; }
+
+    internal IntPtr ReturnRef { get; private init; }
+
+    internal IntPtr UseDataProtectionKeychain { get; private init; }
+
+    internal IntPtr UseOperationPrompt { get; private init; }
+
+    // Security: key attributes.
+    internal IntPtr AttrKeyType { get; private init; }
+
+    internal IntPtr AttrKeySizeInBits { get; private init; }
+
+    internal IntPtr AttrTokenId { get; private init; }
+
+    internal IntPtr AttrIsPermanent { get; private init; }
+
+    internal IntPtr AttrApplicationTag { get; private init; }
+
+    internal IntPtr AttrAccessControl { get; private init; }
+
+    internal IntPtr PrivateKeyAttrs { get; private init; }
+
+    // Security: attribute values.
+    internal IntPtr KeyTypeEcSecPrimeRandom { get; private init; }
+
+    internal IntPtr TokenIdSecureEnclave { get; private init; }
+
+    internal IntPtr AccessibleWhenUnlockedThisDeviceOnly { get; private init; }
+
+    /// <summary>
+    /// <c>kSecKeyAlgorithmECIESEncryptionCofactorX963SHA256AESGCM</c>.
+    /// </summary>
+    /// <remarks>
+    /// The one algorithm the Secure Enclave's P-256 keys support for encryption, and the reason this
+    /// store wraps rather than signs. The long name spells out the whole construction: an ephemeral
+    /// key agreed against the enclave's public half with cofactor ECDH, run through the X9.63 KDF with
+    /// SHA-256, used as an AES-GCM key. The ephemeral public key travels in the output, which is why the
+    /// ciphertext is larger than the 32 bytes going in and why nothing else has to be stored beside it.
+    /// </remarks>
+    internal IntPtr EciesAlgorithm { get; private init; }
+
+    /// <summary>The resolved symbols, loaded once.</summary>
+    internal static MacSymbols Resolve() => Cached.Value;
+
+    private static MacSymbols Load()
+    {
+        try
+        {
+            if (!NativeLibrary.TryLoad(MacSecurity.CoreFoundation, out var cf)
+                || !NativeLibrary.TryLoad(MacSecurity.SecurityFramework, out var sec))
+            {
+                // Every pointer left at zero, which AllResolved reads as incomplete.
+                return new MacSymbols();
+            }
+
+            // The two callback tables are structs rather than object pointers, so what is wanted is the
+            // address of the export itself and not what it holds. Every other symbol here is a CFTypeRef
+            // global and needs the dereference; these two do not, and mixing them up produces a
+            // dictionary that does not retain its contents.
+            var keyCallBacks = NativeLibrary.TryGetExport(cf, "kCFTypeDictionaryKeyCallBacks", out var k)
+                ? k
+                : IntPtr.Zero;
+
+            var valueCallBacks = NativeLibrary.TryGetExport(cf, "kCFTypeDictionaryValueCallBacks", out var v)
+                ? v
+                : IntPtr.Zero;
+
+            return new MacSymbols
+            {
+                True = MacSecurity.Constant(cf, "kCFBooleanTrue"),
+                TypeDictionaryKeyCallBacks = keyCallBacks,
+                TypeDictionaryValueCallBacks = valueCallBacks,
+
+                Class = MacSecurity.Constant(sec, "kSecClass"),
+                ClassKey = MacSecurity.Constant(sec, "kSecClassKey"),
+                ReturnRef = MacSecurity.Constant(sec, "kSecReturnRef"),
+                UseDataProtectionKeychain = MacSecurity.Constant(sec, "kSecUseDataProtectionKeychain"),
+                UseOperationPrompt = MacSecurity.Constant(sec, "kSecUseOperationPrompt"),
+
+                AttrKeyType = MacSecurity.Constant(sec, "kSecAttrKeyType"),
+                AttrKeySizeInBits = MacSecurity.Constant(sec, "kSecAttrKeySizeInBits"),
+                AttrTokenId = MacSecurity.Constant(sec, "kSecAttrTokenID"),
+                AttrIsPermanent = MacSecurity.Constant(sec, "kSecAttrIsPermanent"),
+                AttrApplicationTag = MacSecurity.Constant(sec, "kSecAttrApplicationTag"),
+                AttrAccessControl = MacSecurity.Constant(sec, "kSecAttrAccessControl"),
+                PrivateKeyAttrs = MacSecurity.Constant(sec, "kSecPrivateKeyAttrs"),
+
+                KeyTypeEcSecPrimeRandom = MacSecurity.Constant(sec, "kSecAttrKeyTypeECSECPrimeRandom"),
+                TokenIdSecureEnclave = MacSecurity.Constant(sec, "kSecAttrTokenIDSecureEnclave"),
+                AccessibleWhenUnlockedThisDeviceOnly =
+                    MacSecurity.Constant(sec, "kSecAttrAccessibleWhenUnlockedThisDeviceOnly"),
+
+                EciesAlgorithm = MacSecurity.Constant(
+                    sec,
+                    "kSecKeyAlgorithmECIESEncryptionCofactorX963SHA256AESGCM"),
+            };
+        }
+        catch (Exception exception) when (exception is DllNotFoundException or BadImageFormatException)
+        {
+            return new MacSymbols();
+        }
+    }
+
+    private bool AllResolved() =>
+        True != IntPtr.Zero
+        && TypeDictionaryKeyCallBacks != IntPtr.Zero
+        && TypeDictionaryValueCallBacks != IntPtr.Zero
+        && Class != IntPtr.Zero
+        && ClassKey != IntPtr.Zero
+        && ReturnRef != IntPtr.Zero
+        && UseDataProtectionKeychain != IntPtr.Zero
+        && UseOperationPrompt != IntPtr.Zero
+        && AttrKeyType != IntPtr.Zero
+        && AttrKeySizeInBits != IntPtr.Zero
+        && AttrTokenId != IntPtr.Zero
+        && AttrIsPermanent != IntPtr.Zero
+        && AttrApplicationTag != IntPtr.Zero
+        && AttrAccessControl != IntPtr.Zero
+        && PrivateKeyAttrs != IntPtr.Zero
+        && KeyTypeEcSecPrimeRandom != IntPtr.Zero
+        && TokenIdSecureEnclave != IntPtr.Zero
+        && AccessibleWhenUnlockedThisDeviceOnly != IntPtr.Zero
+        && EciesAlgorithm != IntPtr.Zero;
+}

src/DodoSSH.Client.App/Platform/VelopackUpdateChannel.cs

@@ -31,7 +31,12 @@ internal static class UpdateChannels
     /// </remarks>
     internal static IUpdateChannel ForThisMachine()
     {
-        if (!OperatingSystem.IsWindows())
+        // Two platforms now, and the check is a list rather than a negation for a reason: Linux reaches
+        // this too. Velopack has a Linux path — AppImage — but this repository does not build one, so a
+        // Linux build is a checkout somebody ran, and handing it an UpdateManager would have it poll a
+        // feed carrying nothing it could apply. Naming the platforms that are packaged keeps a future
+        // AppImage an addition here rather than a thing that silently already half-happened.
+        if (!OperatingSystem.IsWindows() && !OperatingSystem.IsMacOS())
         {
             return new UnavailableUpdateChannel();
         }
@@ -55,14 +60,21 @@ internal static class UpdateChannels
 }
 
 /// <summary>
-/// The Windows update channel, backed by Velopack against the project's own forge.
+/// The desktop update channel, backed by Velopack against the project's own forge.
 /// </summary>
 /// <remarks>
 /// <para>
-/// The one file in the repository that names Velopack. It lives beside <c>WindowsDeviceKeyStore</c>
-/// rather than in a project of its own because it is the same kind of thing — a Windows-only
-/// implementation of an interface declared in <c>DodoSSH.Client.Session</c> — and because
-/// <c>DodoSSH.Client.Shell</c> is shared with the Android head, which must never acquire an updater.
+/// The one file in the repository that names Velopack. It lives beside the platform key stores rather
+/// than in a project of its own because it is the same kind of thing — a desktop-only implementation of
+/// an interface declared in <c>DodoSSH.Client.Session</c> — and because <c>DodoSSH.Client.Shell</c> is
+/// shared with the Android head, which must never acquire an updater.
+/// </para>
+/// <para>
+/// <b>One class for both desktop platforms, where the key stores are one class each.</b> The difference
+/// is where the platform knowledge sits. A key store is platform knowledge from top to bottom: different
+/// hardware, different API, different failure modes. Velopack's <c>UpdateManager</c> has already absorbed
+/// all of that, and what is left over — check, download, apply, restart — is identical on the two. The
+/// only thing that differs is which string names the feed, and that is <see cref="ChannelFor"/>.
 /// </para>
 /// <para>
 /// See <c>docs/adr/0013-desktop-distribution-and-updates.md</c>.
@@ -103,7 +115,7 @@ internal sealed class VelopackUpdateChannel : IUpdateChannel
     /// but unsaid on one side and stated on the other is how a feed goes quiet with no error anywhere:
     /// the check succeeds, finds nothing, and reports that the client is up to date forever.
     /// </remarks>
-    private const string ReleaseChannel = "win";
+    private const string WindowsReleaseChannel = "win";
 
     /// <summary>
     /// The nightly channel, which is a different name rather than the same one on a different tag.
@@ -122,7 +134,26 @@ internal sealed class VelopackUpdateChannel : IUpdateChannel
     /// a download somebody watched. A channel each means neither ever sees the other's releases at all.
     /// </para>
     /// </remarks>
-    private const string NightlyChannel = "win-nightly";
+    private const string WindowsNightlyChannel = "win-nightly";
+
+    /// <summary>The macOS release channel, and Velopack's own default there.</summary>
+    /// <remarks>
+    /// A contract with <c>scripts/release-macos.sh</c>, exactly as the Windows pair is one with the
+    /// PowerShell script. Stated for the same reason, which applies with more force here: the four
+    /// channels all publish to one repository, so the only thing keeping a Mac from being offered a
+    /// <c>win</c> package is that it never reads that index.
+    /// </remarks>
+    private const string MacReleaseChannel = "osx";
+
+    /// <summary>The macOS nightly channel.</summary>
+    /// <remarks>
+    /// Named here and not yet published by anything. The CI job for the macOS head builds and bundles
+    /// and deliberately uploads nothing — see the packaging step in <c>ci.yml</c> — so a nightly macOS
+    /// build checking this feed finds an empty channel and reports itself up to date, which is the
+    /// correct behaviour for a channel with no publisher. The name exists so that turning the publisher
+    /// on later is one job rather than a job plus a rename that has to reach every installed client.
+    /// </remarks>
+    private const string MacNightlyChannel = "osx-nightly";
 
     private readonly UpdateManager manager;
 
@@ -141,10 +172,13 @@ internal sealed class VelopackUpdateChannel : IUpdateChannel
     /// <inheritdoc />
     public bool IsSupported => true;
 
-    /// <summary>Always, on this head.</summary>
+    /// <summary>Always, on this head, on either platform.</summary>
     /// <remarks>
-    /// Velopack's apply runs <c>Update.exe</c> over this installation and restarts it, so the process is
-    /// gone by the time anything could have asked a question. The phone's is the other answer; see
+    /// Velopack's apply hands off to a separate updater process — <c>Update.exe</c> on Windows, the
+    /// <c>UpdateMac</c> helper inside the bundle on macOS — which replaces this installation and
+    /// relaunches it, so the process is gone by the time anything could have asked a question. The
+    /// mechanism differs and the answer does not, which is why this is a constant rather than another
+    /// thing <see cref="ChannelFor"/> would have to decide. The phone's is the other answer; see
     /// <see cref="IUpdateChannel.ApplyingEndsTheProcess"/> for what the caller does differently.
     /// </remarks>
     public bool ApplyingEndsTheProcess => true;
@@ -183,7 +217,36 @@ internal sealed class VelopackUpdateChannel : IUpdateChannel
 
         return new UpdateManager(
             new GiteaSource(RepositoryUrl, accessToken: null, prerelease: nightly),
-            new UpdateOptions { ExplicitChannel = nightly ? NightlyChannel : ReleaseChannel });
+            new UpdateOptions { ExplicitChannel = ChannelFor(nightly) });
+    }
+
+    /// <summary>
+    /// The one of the four channel names this build belongs to.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// Two independent axes — which platform, and which of that platform's two channels — and they are
+    /// resolved in one place so that neither can be answered differently somewhere else. The platform
+    /// half is the running OS rather than anything recorded in the build, because a package can only
+    /// ever be applied on the platform it was built for; the channel half comes from assembly metadata,
+    /// because a release build and a nightly are the same bytes on the same OS and only the metadata
+    /// tells them apart.
+    /// </para>
+    /// <para>
+    /// Windows is the fallback rather than a third branch. Only Windows and macOS reach here at all —
+    /// <see cref="UpdateChannels.ForThisMachine"/> is the gate — so the alternative would be an
+    /// unreachable throw, and an unreachable throw in the middle of the updater is a thing somebody
+    /// later has to reason about to discover it cannot happen.
+    /// </para>
+    /// </remarks>
+    private static string ChannelFor(bool nightly)
+    {
+        if (OperatingSystem.IsMacOS())
+        {
+            return nightly ? MacNightlyChannel : MacReleaseChannel;
+        }
+
+        return nightly ? WindowsNightlyChannel : WindowsReleaseChannel;
     }
 
     /// <inheritdoc />

src/DodoSSH.Client.App/Platform/WindowsDeviceKeyStore.cs

@@ -9,9 +9,17 @@ namespace DodoSSH.Client.App.Platform;
 /// </summary>
 /// <remarks>
 /// <para>
-/// One place decides, so nothing above has to carry a platform guard. A machine with no TPM, or one that
-/// is not Windows, gets <see cref="UnavailableDeviceKeyStore"/> and therefore keeps asking for the
-/// passphrase — which is the honest answer rather than a degraded one.
+/// One place decides, so nothing above has to carry a platform guard. A machine with no secure hardware,
+/// or one that is neither Windows nor macOS, gets <see cref="UnavailableDeviceKeyStore"/> and therefore
+/// keeps asking for the passphrase — which is the honest answer rather than a degraded one.
+/// </para>
+/// <para>
+/// <b>Both real stores are asked whether they work rather than told that they do.</b> Each
+/// <c>IsSupported</c> probes by doing the thing — creating a throwaway key and deleting it — because on
+/// both platforms the provider is present and reports itself present on machines where creating a key
+/// fails: a Windows box with no usable TPM, a Mac with no Secure Enclave, and on macOS also every
+/// unsigned development build, since enclave keys need a signing identity. Inferring from the OS would
+/// mean each of those discovering the truth at the moment somebody tried to unlock.
 /// </para>
 /// <para>
 /// <b>"Desktop", because the choice belongs to a head rather than to the session layer.</b> This file used
@@ -29,9 +37,17 @@ public static class DesktopDeviceKeyStores
     {
         ArgumentNullException.ThrowIfNull(paths);
 
-        return OperatingSystem.IsWindows() && WindowsDeviceKeyStore.IsSupported()
-            ? new WindowsDeviceKeyStore(paths)
-            : new UnavailableDeviceKeyStore();
+        if (OperatingSystem.IsWindows() && WindowsDeviceKeyStore.IsSupported())
+        {
+            return new WindowsDeviceKeyStore(paths);
+        }
+
+        if (OperatingSystem.IsMacOS() && MacDeviceKeyStore.IsSupported())
+        {
+            return new MacDeviceKeyStore(paths);
+        }
+
+        return new UnavailableDeviceKeyStore();
     }
 }
 

src/DodoSSH.Client.App/Views/StatusBar.axaml

@@ -25,6 +25,14 @@
     channel this application has for saying that a save failed, that a sync was refused, that a merge picked
     a winner. The design is a mock-up of a working afternoon and has nowhere for a sentence like that to go;
     dropping the bar would have meant dropping the sentence, or repeating it on six screens.
+
+    ── One line, several terminals ────────────────────────────────────────────────────────────────────────
+    One bar for a window that now holds any number of tabs, and connecting writes to it — so "Connected to
+    prod-db." outlived the tab it was about, and closing prod-db left that sentence sitting under whichever
+    terminal the user looked at next. A connection's own lines are now owned by the attempt that wrote them
+    and are cleared when its tab is closed; everything else the vault says stays, because closing a terminal
+    is not an answer to a failed save. See VaultViewModel.ForgetConnectionStatus and
+    MainWindowViewModel.CloseTabAsync.
   -->
 
   <Border Height="24" Background="{StaticResource Chrome}"

src/DodoSSH.Client.Shell/ViewModels/MainWindowViewModel.cs

@@ -525,6 +525,7 @@ internal sealed partial class MainWindowViewModel : ObservableObject, IAsyncDisp
         // list. Detached in DisposeAsync, which is the only point either of them ends.
         this.workspace.SessionEnded += OnWorkspaceSessionEnded;
         this.workspace.FontSizeStepRequested += OnFontSizeStepRequested;
+        this.workspace.RendererReattached += OnRendererReattached;
 
         settings = new ClientSettingsStore(paths);
 
@@ -635,6 +636,31 @@ internal sealed partial class MainWindowViewModel : ObservableObject, IAsyncDisp
     private void OnFontSizeStepRequested(object? sender, TerminalFontSizeStepEventArgs e) =>
         Dispatcher.UIThread.Post(() => StepTerminalFontSize(e.Step));
 
+    /// <remarks>
+    /// <para>
+    /// Marshalled for the same reason as <see cref="OnFontSizeStepRequested"/>: this arrives on the data
+    /// plane's socket-accept thread, and both properties it reads here — <see cref="TerminalFontSize"/> and
+    /// <see cref="SelectedTab"/> — are bound to by the interface.
+    /// </para>
+    /// <para>
+    /// <see cref="TerminalWorkspace.RendererReattached"/> fires once the workspace has replayed what it
+    /// owns — the live sessions. Font size and the choice of active tab are not the workspace's to know;
+    /// they live here, so this is the other half of putting a reattached page back the way it was. The size
+    /// is sent exactly as <see cref="TellRendererTheFontSizeAsync"/> sends it at startup, because nothing
+    /// has changed — the page has merely forgotten, and this is only a reminder.
+    /// </para>
+    /// </remarks>
+    private void OnRendererReattached(object? sender, EventArgs e) =>
+        Dispatcher.UIThread.Post(() =>
+        {
+            _ = workspace.SetFontSizeAsync(TerminalFontSize, CancellationToken.None).AsTask();
+
+            if (SelectedTab is { } tab)
+            {
+                _ = workspace.ActivateSessionAsync(tab.SessionId, CancellationToken.None).AsTask();
+            }
+        });
+
     [ObservableProperty]
     private ShellState state = ShellState.Starting;
 
@@ -1898,9 +1924,18 @@ internal sealed partial class MainWindowViewModel : ObservableObject, IAsyncDisp
     /// Closes one terminal, ending its shell.
     /// </summary>
     /// <remarks>
+    /// <para>
     /// This is the one thing in the application that deliberately ends a session, which is why it is a tab's
     /// close button and not a menu item: closing the window somebody's job is running in should take exactly
     /// as much intent as it looks like it does. Locking does not do this, and neither does anything else.
+    /// </para>
+    /// <para>
+    /// <b>It takes the status line with it, when that line is this tab's.</b> The strip along the bottom is
+    /// one line for the whole window — see <c>StatusBar.axaml</c> — and connecting writes to it, so closing
+    /// prod-db used to leave "Connected to prod-db." reporting on a session that is gone, over whichever
+    /// terminal the user looked at next. Only the sentence this attempt wrote goes; anything the vault has
+    /// said since stays, because a failed save is not something closing a terminal answers.
+    /// </para>
     /// </remarks>
     [RelayCommand]
     private async Task CloseTabAsync(TerminalTabViewModel tab)
@@ -1910,9 +1945,9 @@ internal sealed partial class MainWindowViewModel : ObservableObject, IAsyncDisp
             return;
         }
 
-        // Removed first, so the workspace's SessionEnded — which fires as the pump unwinds — finds no tab to
-        // mark dead and does nothing. The alternative ordering leaves a window in which a tab that is on its
-        // way out is repainted as disconnected.
+        // Removed first, so the workspace's SessionEnded — announced once the close below has fully drained
+        // — finds no tab to mark dead and does nothing here. The alternative ordering leaves a window in
+        // which a tab that is on its way out is repainted as disconnected.
         var index = Tabs.IndexOf(tab);
         Tabs.Remove(tab);
 
@@ -1934,6 +1969,9 @@ internal sealed partial class MainWindowViewModel : ObservableObject, IAsyncDisp
 
         RaiseTabState();
 
+        // Only this tab's own sentence, and only if that is what the bar currently holds — see the remark.
+        Vault?.ForgetConnectionStatus(tab.AttemptId);
+
         // Explicitly, and not left to the selection having moved. Closing a tab that was not the selected one
         // changes nothing about the selection, so OnSelectedTabChanged does not run — and the host whose
         // terminal just went would keep a lit dot until something else happened to move the selection.
@@ -3051,6 +3089,7 @@ internal sealed partial class MainWindowViewModel : ObservableObject, IAsyncDisp
 
         workspace.SessionEnded -= OnWorkspaceSessionEnded;
         workspace.FontSizeStepRequested -= OnFontSizeStepRequested;
+        workspace.RendererReattached -= OnRendererReattached;
         transfers.PropertyChanged -= OnTransfersPropertyChanged;
 
         // Stopped here rather than left to the process exiting with it: the loop holds no vault key and
@@ -3373,7 +3412,7 @@ internal sealed partial class MainWindowViewModel : ObservableObject, IAsyncDisp
     /// </remarks>
     private void OnVaultConnectionStarting(object? sender, ConnectionAttemptEventArgs e)
     {
-        var tab = new TerminalTabViewModel(e.Label, e.Address);
+        var tab = new TerminalTabViewModel(e.Label, e.Address) { AttemptId = e.AttemptId };
         attempts[e.AttemptId] = tab;
 
         AdoptTab(tab);
@@ -3402,6 +3441,7 @@ internal sealed partial class MainWindowViewModel : ObservableObject, IAsyncDisp
             // dropping it would leave a shell running with nothing in the window naming it.
             var adopted = new TerminalTabViewModel(e.SessionId, e.Label, e.Address)
             {
+                AttemptId = e.AttemptId,
                 StartedAt = clock.GetUtcNow(),
                 Cipher = NullIfEmpty(e.Cipher),
                 HostKeyAlgorithm = NullIfEmpty(e.HostKeyAlgorithm),

src/DodoSSH.Client.Shell/ViewModels/TerminalTabViewModel.cs

@@ -84,6 +84,18 @@ internal sealed partial class TerminalTabViewModel : ObservableObject
     /// </remarks>
     internal uint SessionId { get; private set; }
 
+    /// <summary>
+    /// The connection attempt this tab was opened for, or <see cref="Guid.Empty"/> for a tab nothing dialled.
+    /// </summary>
+    /// <remarks>
+    /// Carried only so that closing this tab can take the vault's status line with it when the line is about
+    /// this connection — see <c>VaultViewModel.ForgetConnectionStatus</c>. It is the attempt id and not the
+    /// session id because the message can be written before a session exists ("Connecting to prod-db…") and
+    /// after one has stopped existing, and because the two failure sentences a tab can end on never have a
+    /// session id at all.
+    /// </remarks>
+    internal Guid AttemptId { get; init; }
+
     internal string Label { get; }
 
     /// <summary>The account and endpoint, for the pane header and the status bar.</summary>

src/DodoSSH.Client.Shell/ViewModels/TransfersViewModel.cs

@@ -620,6 +620,19 @@ internal sealed partial class TransfersViewModel : ObservableObject, IAsyncDispo
     [ObservableProperty]
     private string? connectedCipher;
 
+    /// <summary>
+    /// Whether there is a live SFTP connection this session would lose by dying — the phone's foreground-
+    /// service question, not the desktop's.
+    /// </summary>
+    /// <remarks>
+    /// <see cref="ConnectedCipher"/> is already the fact that tells a host apart from a bucket, because only
+    /// a host set it — a bucket is HTTP, per-request, and closes nothing a dying process would have kept
+    /// open, so it answers false here even while <see cref="IsConnected"/> is true. Android reads this to
+    /// decide whether an idle Files screen with no transfer moving still needs the process kept alive; the
+    /// desktop has no such question because nothing stops its process for having gone quiet.
+    /// </remarks>
+    internal bool HasLiveFileSession => IsConnected && ConnectedCipher is not null;
+
     /// <summary>The accepted host key's algorithm, e.g. <c>ssh-ed25519</c>. See <see cref="ConnectedCipher"/>.</summary>
     [ObservableProperty]
     private string? connectedHostKeyAlgorithm;
@@ -637,6 +650,31 @@ internal sealed partial class TransfersViewModel : ObservableObject, IAsyncDispo
     [ObservableProperty]
     private string? connectedIdentityLabel;
 
+    /// <summary>
+    /// The paths pinned on the connected host, for the phone's Files-screen chip row — the desktop draws
+    /// the same list in its QUICK ACCESS sidebar, over the terminal surface rather than this one. See
+    /// <see cref="VaultViewModel.EditorPinnedPaths"/> for where a pin is actually added or removed; this is
+    /// a read of what was already saved there.
+    /// </summary>
+    /// <remarks>
+    /// Captured at connect, the same moment <see cref="ConnectedTo"/> is, rather than followed live off the
+    /// host row's own <c>PinnedPaths</c>. A pin edited while this session stays open shows up on the next
+    /// connect rather than mid-session — the same lag <see cref="ConnectedTo"/> itself already carries for
+    /// a relabel — because this screen reads the vault once, at the moment it dials, rather than staying
+    /// wired to a collection it otherwise never has a reason to watch. A bucket has no pins at all:
+    /// <see cref="OpenBucketAsync"/> leaves this empty rather than reading as "not yet known", which is what
+    /// empty already means for a host that connected with none pinned.
+    /// </remarks>
+    internal ObservableCollection<string> ConnectedPinnedPaths { get; } = [];
+
+    /// <summary>Whether the connected host or bucket has any pins to draw as chips.</summary>
+    /// <remarks>
+    /// A read of <see cref="ConnectedPinnedPaths"/> rather than an <c>[ObservableProperty]</c> of its own,
+    /// so it is raised by hand at each of the three places that collection is repopulated or cleared —
+    /// <see cref="MarkHostConnected"/>, <see cref="OpenBucketAsync"/> and <see cref="CloseSessionAsync"/>.
+    /// </remarks>
+    internal bool HasConnectedPins => ConnectedPinnedPaths.Count > 0;
+
     [ObservableProperty]
     private HostKeyPresentation? pendingHostKey;
 
@@ -720,13 +758,18 @@ internal sealed partial class TransfersViewModel : ObservableObject, IAsyncDispo
 
     internal ObservableCollection<TransferRowViewModel> Transfers { get; } = [];
 
-    /// <summary>Raised on the UI thread whenever a transfer appears or changes state.</summary>
+    /// <summary>
+    /// Raised on the UI thread whenever a transfer appears or changes state, or a host or bucket connects or
+    /// disconnects.
+    /// </summary>
     /// <remarks>
     /// For a head that has to tell the operating system what this process is doing — Android's foreground
-    /// service, which must be up for as long as bytes are moving and down afterwards. An event rather than
-    /// letting that head watch <see cref="Transfers"/> itself: the collection announces rows arriving and
-    /// leaving, and the transition that matters most is neither of those but a row going from RUNNING to
-    /// DONE without moving.
+    /// service, which must be up for as long as bytes are moving, or a host session sits open, and down
+    /// afterwards. An event rather than letting that head watch <see cref="Transfers"/> itself: the
+    /// collection announces rows arriving and leaving, and the transition that matters most is neither of
+    /// those but a row going from RUNNING to DONE without moving. Connecting and disconnecting are the other
+    /// two transitions the service cares about — see <see cref="HasLiveFileSession"/> — and neither touches
+    /// <see cref="Transfers"/> at all, so they need this same announcement made by hand.
     /// </remarks>
     internal event EventHandler? ActivityChanged;
 
@@ -947,6 +990,10 @@ internal sealed partial class TransfersViewModel : ObservableObject, IAsyncDispo
                 ConnectedHostKeyAlgorithm = null;
                 ConnectedIdentityLabel = null;
 
+                // And no pins either — see ConnectedPinnedPaths's own remark.
+                ConnectedPinnedPaths.Clear();
+                OnPropertyChanged(nameof(HasConnectedPins));
+
                 connected = (ConnectedTo, row.Label, row.EntityId, TimeProvider.System.GetUtcNow());
 
                 await ListRemoteAsync(session.HomeDirectory, cancellationToken).ConfigureAwait(true);
@@ -1033,10 +1080,26 @@ internal sealed partial class TransfersViewModel : ObservableObject, IAsyncDispo
         ConnectedHostKeyAlgorithm = opened.HostKey.Algorithm;
         ConnectedIdentityLabel = identityLabel;
 
+        // See ConnectedPinnedPaths's own remark for why this is a snapshot rather than a live follow.
+        ConnectedPinnedPaths.Clear();
+        foreach (var path in row.Host.PinnedPaths)
+        {
+            ConnectedPinnedPaths.Add(path);
+        }
+
+        OnPropertyChanged(nameof(HasConnectedPins));
+
         // Recorded, and not hidden because it is "only" the file browser. Opening this is a second login as
         // far as the remote's own auth.log is concerned, so a log of ours that omitted it would disagree with
         // the host's — and anybody comparing the two would be right to believe the host.
         connected = (ConnectedTo, row.Label, row.EntityId, TimeProvider.System.GetUtcNow());
+
+        // Raised here rather than from OnIsConnectedChanged, on purpose: IsConnected is set first, above,
+        // and ConnectedCipher second — a partial method firing off the first assignment would read
+        // HasLiveFileSession against a ConnectedCipher still holding whatever the previous session left
+        // there. Only at the end of this method are both facts actually true together.
+        OnPropertyChanged(nameof(HasLiveFileSession));
+        ActivityChanged?.Invoke(this, EventArgs.Empty);
     }
 
     /// <summary>
@@ -1811,11 +1874,18 @@ internal sealed partial class TransfersViewModel : ObservableObject, IAsyncDispo
         ConnectedCipher = null;
         ConnectedHostKeyAlgorithm = null;
         ConnectedIdentityLabel = null;
+        ConnectedPinnedPaths.Clear();
+        OnPropertyChanged(nameof(HasConnectedPins));
         RemotePath = string.Empty;
         RemoteEntries.Clear();
         RemoteTrail.Clear();
         SelectedRemoteEntry = null;
 
+        // Same ordering reason as the raise at the end of MarkHostConnected: both properties this reads are
+        // already null above, so the raise belongs after them rather than in OnIsConnectedChanged. This also
+        // covers OpenBucketAsync, which calls this method first and never itself turns HasLiveFileSession on.
+        OnPropertyChanged(nameof(HasLiveFileSession));
+        ActivityChanged?.Invoke(this, EventArgs.Empty);
     }
 
     /// <remarks>

src/DodoSSH.Client.Shell/ViewModels/VaultViewModel.cs

@@ -2130,6 +2130,60 @@ internal sealed partial class VaultViewModel(
     [ObservableProperty]
     private string status = string.Empty;
 
+    /// <summary>
+    /// The connection <see cref="Status"/> is currently about, or <see cref="Guid.Empty"/> when it is about
+    /// something else — a save, a sync, a refusal that has nothing to do with a terminal.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// The status line is one line for the whole window, and most of what goes through it is about the vault,
+    /// which is one thing however many terminals are open. A connection is not: "Connected to prod-db." is
+    /// about one tab, and it used to outlive that tab — close prod-db, look at the terminal beside it, and the
+    /// strip along the bottom was still reporting on a session that is no longer there.
+    /// </para>
+    /// <para>
+    /// Ownership rather than a blanket clear on every close, because the alternative loses the sentences this
+    /// bar exists for: a failed save or a refused sync is what the design deleted the bar's other four fields
+    /// to make room for, and closing a terminal is no reason to take one away.
+    /// </para>
+    /// <para>
+    /// Reset by <see cref="OnStatusChanged"/> on every write, so the only messages that carry an owner are the
+    /// ones <see cref="SayAbout"/> writes. Anything assigning <see cref="Status"/> in the ordinary way says,
+    /// by doing so, that its message belongs to no tab.
+    /// </para>
+    /// </remarks>
+    private Guid statusAttemptId;
+
+    partial void OnStatusChanged(string value) => statusAttemptId = Guid.Empty;
+
+    /// <summary>Writes a status line that belongs to one connection attempt.</summary>
+    /// <remarks>
+    /// After the assignment, not before: <see cref="OnStatusChanged"/> clears the owner on every write, which
+    /// is what makes "belongs to nobody" the default rather than something each of the other status writes
+    /// would have to remember to say.
+    /// </remarks>
+    private void SayAbout(ConnectionAttemptEventArgs attempt, string message)
+    {
+        Status = message;
+        statusAttemptId = attempt.AttemptId;
+    }
+
+    /// <summary>
+    /// Clears the status line if what it is saying is about this connection, and leaves it alone otherwise.
+    /// </summary>
+    /// <remarks>
+    /// Called by the shell when a tab is closed — see <c>MainWindowViewModel.CloseTabAsync</c>. Cleared to
+    /// empty rather than replaced with "Closed prod-db.": the tab going is the report, it is the report the
+    /// user just made happen, and a bar that answers every close with a sentence is a bar people stop reading.
+    /// </remarks>
+    internal void ForgetConnectionStatus(Guid attemptId)
+    {
+        if (attemptId != Guid.Empty && statusAttemptId == attemptId)
+        {
+            Status = string.Empty;
+        }
+    }
+
     [ObservableProperty]
     private int pendingChanges;
 
@@ -10767,7 +10821,7 @@ internal sealed partial class VaultViewModel(
             Dialled(target, authentication));
 
         ConnectionStarting?.Invoke(this, attempt);
-        Status = $"Connecting to {target.Label}…";
+        SayAbout(attempt, $"Connecting to {target.Label}…");
 
         await OpenSessionAsync(attempt, target, authentication, cancellationToken).ConfigureAwait(true);
     }
@@ -11042,7 +11096,9 @@ internal sealed partial class VaultViewModel(
     /// </remarks>
     private void Abandon(ConnectionAttemptEventArgs attempt, string reason)
     {
-        Status = reason;
+        // Owned by the attempt, unlike Answer's own line below: this is the case where the tab stays, carrying
+        // the same reason, so closing that tab is exactly the moment the line stops being about anything.
+        SayAbout(attempt, reason);
 
         ConnectionFailed?.Invoke(
             this,
@@ -11097,7 +11153,7 @@ internal sealed partial class VaultViewModel(
         // only record that machine was reached at all.
         connectionLog?.Identify(sessionId, target.Label, target.HostId);
 
-        Status = $"Connected to {target.Label}.";
+        SayAbout(attempt, $"Connected to {target.Label}.");
 
         // The session the workspace just opened is the only place the negotiated cipher and host-key
         // algorithm live — an SshConnectionRequest asks for neither and gets no say in either — so they are
@@ -11128,7 +11184,8 @@ internal sealed partial class VaultViewModel(
         // screen it is typed on says so.
         if (target.Row is { } row)
         {
-            await RememberTypedPasswordAsync(row, authentication, cancellationToken).ConfigureAwait(true);
+            await RememberTypedPasswordAsync(attempt, row, authentication, cancellationToken)
+                .ConfigureAwait(true);
         }
     }
 
@@ -11160,6 +11217,7 @@ internal sealed partial class VaultViewModel(
     /// </para>
     /// </remarks>
     private async Task RememberTypedPasswordAsync(
+        ConnectionAttemptEventArgs attempt,
         HostRowViewModel row,
         HostAuthentication authentication,
         CancellationToken cancellationToken)
@@ -11175,8 +11233,10 @@ internal sealed partial class VaultViewModel(
 
         if (row.IsReadOnly)
         {
-            Status = $"Connected to {row.Label}. Its password was not saved: this host was written by a "
-                + "newer version of DodoSSH, and binding a credential would re-encode it.";
+            SayAbout(
+                attempt,
+                $"Connected to {row.Label}. Its password was not saved: this host was written by a newer "
+                + "version of DodoSSH, and binding a credential would re-encode it.");
             return;
         }
 
@@ -11200,7 +11260,9 @@ internal sealed partial class VaultViewModel(
         }
         catch (Exception exception)
         {
-            Status = $"Connected to {row.Label}, but its password could not be saved: {exception.Message}";
+            SayAbout(
+                attempt,
+                $"Connected to {row.Label}, but its password could not be saved: {exception.Message}");
             return;
         }
 
@@ -11211,8 +11273,10 @@ internal sealed partial class VaultViewModel(
 
         await ReloadAsync(cancellationToken).ConfigureAwait(true);
 
-        Status = $"Connected to {row.Label}. Its password is saved in your keychain as '{row.Label}', so it "
-            + "will not be asked for again.";
+        SayAbout(
+            attempt,
+            $"Connected to {row.Label}. Its password is saved in your keychain as '{row.Label}', so it will "
+            + "not be asked for again.");
 
         await AutoSyncAsync(cancellationToken).ConfigureAwait(true);
     }

src/DodoSSH.Client.Shell/WebAssets/terminal.js

@@ -3,7 +3,7 @@
 /*
   The renderer half of the terminal data plane.
 
-  Two things here are load-bearing and easy to get wrong:
+  Three things here are load-bearing and easy to get wrong:
 
   1. Output is acknowledged from term.write's completion callback, never on receipt. The
      acknowledgement returns flow-control credit to the host, so acknowledging early would tell
@@ -15,6 +15,14 @@
      partial sequences across writes. Decoding here would corrupt any multi-byte character that
      happened to straddle a frame boundary, which shows up as occasional mojibake in exactly the
      conditions that are hardest to reproduce.
+
+  3. The socket reconnects itself, forever, with backoff. This page's WebView is routinely killed
+     and reloaded by Android under memory pressure or simply for being backgrounded, so "the
+     socket closed" is an ordinary event here, not the end of the terminal's life — see connect().
+     A reloaded page starts with an empty session map, so createSession is idempotent (a session
+     that already has a pane is left alone) and a SESSION_OPENED frame carries a flag telling this
+     page whether it is a replay: nothing to do for a pane that is still here, and a short banner
+     for one that is not, because that pane's scrollback genuinely did not survive.
 */
 
 const SERVER_OUTPUT = 1;
@@ -33,6 +41,25 @@ const CLIENT_FONT_SIZE_STEP = 4;
 const HEADER_LENGTH = 5;
 const SCROLLBACK_LINES = 5000;
 
+/*
+  How long to wait before trying the socket again, and how that wait grows. Starting quick matters
+  because the ordinary case is a page that just finished loading after its WebView came back — the
+  host's listener has been sitting there the whole time — and capping it matters because there is no
+  point spacing attempts further apart than a person notices. Forever rather than giving up, because
+  giving up would need a way to try again and there is none better than the one already here: the page
+  dies with the app.
+*/
+const RECONNECT_INITIAL_DELAY_MS = 1000;
+const RECONNECT_MAX_DELAY_MS = 5000;
+
+/*
+  Styled like the SESSION_CLOSED banner (matching \x1b[38;5;244, the same dim grey), but written by
+  createSession's caller rather than by createSession itself: only a *replay* landing on a pane that
+  does not exist yet means the page reloaded and lost it, and createSession has no way to know which
+  of its callers that is.
+*/
+const REPLAY_BANNER = '\x1b[38;5;244m── the view reconnected; earlier output stayed on the host ──\x1b[0m\r\n';
+
 /*
   The size panes are created at, until the host says otherwise — which it does as soon as it has read
   the stored preference, usually before the first session exists. Kept here as well so a pane opened
@@ -192,7 +219,19 @@ function handleKey(event) {
   return true;
 }
 
+/**
+ * Builds a pane for a session, or returns the one already there.
+ *
+ * Idempotent because a replay can land on a page that never lost its pane — the socket dropped and
+ * came back, but this page's own process survived — and asking for a session that already has a pane
+ * must not build a second one on top of it, orphaning the first one's WebGL context and scrollback.
+ */
 function createSession(sessionId) {
+  const existing = sessions.get(sessionId);
+  if (existing) {
+    return existing;
+  }
+
   const pane = document.createElement('div');
   pane.className = 'pane';
   pane.dataset.sessionId = String(sessionId);
@@ -290,10 +329,23 @@ function handleFrame(buffer) {
   const payload = new Uint8Array(buffer, HEADER_LENGTH);
 
   switch (opcode) {
-    case SERVER_SESSION_OPENED:
-      createSession(sessionId);
+    case SERVER_SESSION_OPENED: {
+      // Checked before createSession, which would otherwise erase the answer by creating the pane
+      // this check is asking about.
+      const hadPaneAlready = sessions.has(sessionId);
+      const session = createSession(sessionId);
+
+      // Byte 1 means the host is replaying a session that existed before this socket attached — see
+      // TerminalWorkspace.ReplayAfterAttachAsync. A replay landing on a pane that is still here has
+      // nothing left to do beyond the idempotent create above; one landing on a pane that is not means
+      // this page reloaded and that pane's scrollback went with it, which is worth a line saying so.
+      if (payload.length > 0 && payload[0] === 1 && !hadPaneAlready) {
+        session.term.write(REPLAY_BANNER);
+      }
+
       setStatus('');
       break;
+    }
 
     case SERVER_OUTPUT: {
       const session = sessions.get(sessionId) ?? createSession(sessionId);
@@ -414,6 +466,31 @@ function handleFrame(buffer) {
   }
 }
 
+/** @type {number | null} */
+let reconnectTimer = null;
+let reconnectDelay = RECONNECT_INITIAL_DELAY_MS;
+
+/**
+ * Tries the socket again after a wait, unless a try is already pending.
+ *
+ * The guard is what keeps 'close' and 'error' from stacking two timers for one failure — a socket
+ * that fails to open typically fires both, and each would otherwise schedule its own reconnect.
+ */
+function scheduleReconnect() {
+  if (reconnectTimer !== null) {
+    return;
+  }
+
+  setStatus('Reconnecting the terminal view…');
+
+  reconnectTimer = setTimeout(() => {
+    reconnectTimer = null;
+    connect();
+  }, reconnectDelay);
+
+  reconnectDelay = Math.min(reconnectDelay * 2, RECONNECT_MAX_DELAY_MS);
+}
+
 function connect() {
   const token = root.dataset.token;
   const url = root.dataset.socket;
@@ -423,16 +500,22 @@ function connect() {
   socket = new WebSocket(url, ['dodossh.terminal.v1', `token.${token}`]);
   socket.binaryType = 'arraybuffer';
 
-  socket.addEventListener('open', () => setStatus(''));
+  socket.addEventListener('open', () => {
+    setStatus('');
+
+    // Back to the quick attempt for whatever the next failure turns out to be. Kept slow between
+    // attempts within one outage, reset once the outage is actually over.
+    reconnectDelay = RECONNECT_INITIAL_DELAY_MS;
+  });
+
   socket.addEventListener('message', (event) => handleFrame(event.data));
 
-  socket.addEventListener('close', () => {
-    setStatus('Disconnected from DodoSSH.');
-  });
-
-  socket.addEventListener('error', () => {
-    setStatus('The terminal connection failed.');
-  });
+  // Both close and error retry. They are not the same event on every failure — a socket that never
+  // opens can fire only 'error', one that opens and later drops fires only 'close' — and the host
+  // side of this same problem (TerminalDataPlane.UpgradeAsync's takeover) is exactly why retrying is
+  // safe: whichever attempt eventually reaches the host, a fresh valid upgrade always wins the socket.
+  socket.addEventListener('close', scheduleReconnect);
+  socket.addEventListener('error', scheduleReconnect);
 }
 
 // One observer for the whole root rather than one per pane: resizes arrive in bursts while a

src/DodoSSH.Client.Terminal/TerminalDataPlane.cs

@@ -66,7 +66,6 @@ public sealed class TerminalDataPlane : ITerminalTransport, IAsyncDisposable
         new(TaskCreationOptions.RunContinuationsAsynchronously);
 
     private WebSocket? socket;
-    private int accepted;
     private int disposed;
 
     /// <param name="assets">Where the renderer's files come from.</param>
@@ -106,6 +105,26 @@ public sealed class TerminalDataPlane : ITerminalTransport, IAsyncDisposable
     /// </remarks>
     public event EventHandler<TerminalFontSizeStepEventArgs>? FontSizeStepRequested;
 
+    /// <summary>
+    /// Raised after a socket attaches — the first one, and every later takeover.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// Raised after <see cref="socket"/> has been swapped in but before <see cref="ReceiveLoopAsync"/> starts
+    /// consuming it, on the socket-accept thread — the same thread that is in the middle of
+    /// <see cref="UpgradeAsync"/> for this connection. <see cref="TerminalWorkspace"/> is this event's one
+    /// subscriber, and it uses the ordering to replay session state before anything the fresh page sends
+    /// (a resize, an early acknowledgement) can be dispatched; see its remark for why the two racing is
+    /// harmless regardless.
+    /// </para>
+    /// <para>
+    /// Unlike <see cref="RendererAttached"/>, which resolves once and answers "has a renderer ever attached"
+    /// for <see cref="TerminalWorkspace.WaitForRendererAsync"/>, this fires every time — because a takeover
+    /// is exactly the case <see cref="RendererAttached"/> was never meant to describe again.
+    /// </para>
+    /// </remarks>
+    public event EventHandler? SocketAttached;
+
     /// <summary>Registers a session so inbound frames can be routed to it.</summary>
     public void Register(uint sessionId, TerminalSessionPump pump)
     {
@@ -148,8 +167,9 @@ public sealed class TerminalDataPlane : ITerminalTransport, IAsyncDisposable
             // Each connection on its own task, and deliberately not awaited. An upgraded WebSocket
             // lives for the whole session, so handling connections in sequence would leave the accept
             // loop parked inside the receive loop and every later request unanswered — the page's
-            // script and stylesheet among them. Concurrency needs no coordination here because the
-            // single-attach guard is an interlocked exchange.
+            // script and stylesheet among them. Two upgrades racing each other need no coordination
+            // here either, because the takeover in UpgradeAsync swaps the shared socket field with an
+            // interlocked exchange rather than assuming it is the only writer.
             _ = HandleConnectionAsync(client, linked.Token);
         }
     }
@@ -192,6 +212,29 @@ public sealed class TerminalDataPlane : ITerminalTransport, IAsyncDisposable
                 .SendAsync(frame, WebSocketMessageType.Binary, endOfMessage: true, cancellationToken)
                 .ConfigureAwait(false);
         }
+        catch (OperationCanceledException) when (!cancellationToken.IsCancellationRequested)
+        {
+            // Not a cancellation despite the type: .NET's ManagedWebSocket wraps a send that fails because
+            // the underlying connection is already gone — which is exactly what a killed renderer's socket
+            // looks like — in an OperationCanceledException of its own manufacture, regardless of whether
+            // anyone actually cancelled anything. The filter is what tells the two apart: if the caller's
+            // own token were the cause, IsCancellationRequested would be true here and this catch does not
+            // apply, so a real cancellation still propagates. Everything below about why this must not
+            // fault the caller applies here exactly as it does to the exception types in the next catch.
+        }
+        catch (Exception exception)
+            when (exception is WebSocketException or ObjectDisposedException
+                or InvalidOperationException or IOException)
+        {
+            // The state check above is not atomic with the send, and a WebView renderer process killed by
+            // Android leaves its socket reporting Open long after nobody is reading from the other end. This
+            // has to read as "nobody listening" — the same as the no-socket case above — and never as a
+            // fault: SendAsync is called from TerminalSessionPump.SendOutputAsync inside the flush loop, and
+            // letting this exception escape would fault that loop. A faulted flush loop stops draining the
+            // credit window, the reader blocks once it fills, and the SSH session behind it freezes for good
+            // while LiveSessionCount still counts it as running. A dropped frame is recoverable — a frozen
+            // session is not.
+        }
         finally
         {
             sendGate.Release();
@@ -267,6 +310,25 @@ public sealed class TerminalDataPlane : ITerminalTransport, IAsyncDisposable
             .ConfigureAwait(false);
     }
 
+    /// <remarks>
+    /// <para>
+    /// <b>Takeover, not rejection.</b> A valid upgrade always wins the socket, even when one is already
+    /// attached — the old socket is aborted and the newcomer takes its place. Refusing a second attach used
+    /// to be the rule, on the theory that one renderer lives for the whole process. That is WebView2's
+    /// truth and not Android's: the platform kills the WebView's renderer process under memory pressure or
+    /// simply for being backgrounded, the page reloads, and the reload's socket is a second valid upgrade —
+    /// refusing it left the terminal permanently unreachable with no way back short of restarting the app.
+    /// </para>
+    /// <para>
+    /// Refusing protects nothing here anyway: only our own page knows the token (see the type-level remark
+    /// on what the token defends against), so a second valid upgrade <em>is</em> our page, reattaching.
+    /// Waiting for the old socket to notice it is dead and close on its own is not a safer alternative
+    /// either — a killed renderer process sends no TCP FIN, so the old receive loop can sit unaware for the
+    /// whole 30-second keepalive interval, and every reload landing in that window would still find the
+    /// door held shut by a socket nobody is on the other end of. Taking over immediately is what makes a
+    /// reload actually reattach.
+    /// </para>
+    /// </remarks>
     private async Task UpgradeAsync(
         Stream stream,
         HttpRequestLine request,
@@ -280,16 +342,6 @@ public sealed class TerminalDataPlane : ITerminalTransport, IAsyncDisposable
             return;
         }
 
-        if (Interlocked.Exchange(ref accepted, 1) == 1)
-        {
-            // One renderer, one socket. A second attach would be either a bug or something else on the
-            // machine having found the port.
-            await WriteResponseAsync(
-                    stream, "409 Conflict", "text/plain", "Already attached"u8.ToArray(), cancellationToken)
-                .ConfigureAwait(false);
-            return;
-        }
-
         var key = request.Headers.GetValueOrDefault("sec-websocket-key")!;
         var accept = ComputeHandshakeAccept(key);
 
@@ -314,10 +366,28 @@ public sealed class TerminalDataPlane : ITerminalTransport, IAsyncDisposable
                 KeepAliveInterval = TimeSpan.FromSeconds(30),
             });
 
-        socket = webSocket;
-        rendererAttached.TrySetResult();
+        // Whatever was attached before is displaced, not merely overwritten: Exchange hands back the old
+        // reference so it can be aborted rather than left to linger as a socket nothing reads from again.
+        // Abort rather than a graceful close — a close frame would wait on a peer that, per the remark
+        // above, may never notice it should reply, and the newcomer already proved it is our page.
+        var previous = Interlocked.Exchange(ref socket, webSocket);
+        previous?.Abort();
 
-        await ReceiveLoopAsync(webSocket, cancellationToken).ConfigureAwait(false);
+        rendererAttached.TrySetResult();
+        SocketAttached?.Invoke(this, EventArgs.Empty);
+
+        try
+        {
+            await ReceiveLoopAsync(webSocket, cancellationToken).ConfigureAwait(false);
+        }
+        finally
+        {
+            // Cleared only if the field still holds this connection's own socket. A takeover has already
+            // swapped in a newer one by the time an aborted receive loop unwinds to here, and clearing the
+            // field regardless would race the newcomer: whichever of the two finished last would win, and
+            // it must always be the newcomer, never this one going away.
+            Interlocked.CompareExchange(ref socket, null, webSocket);
+        }
     }
 
     /// <remarks>

src/DodoSSH.Client.Terminal/TerminalWorkspace.cs

@@ -101,6 +101,13 @@ public sealed class TerminalWorkspace : IAsyncDisposable
     private readonly Lock sessionGate = new();
     private readonly CancellationTokenSource lifetime = new();
 
+    /// <summary>
+    /// The payload that marks a <see cref="TerminalServerOpcode.SessionOpened"/> frame as a replay rather
+    /// than a fresh open. A one-byte non-empty payload, so terminal.js's existing length check (empty
+    /// payload for a real open) tells the two apart without a second opcode.
+    /// </summary>
+    private static readonly byte[] ReplayMarker = [1];
+
     private uint nextSessionId = 1;
     private Task? server;
     private int disposed;
@@ -125,6 +132,12 @@ public sealed class TerminalWorkspace : IAsyncDisposable
         // came from. Nothing here decides anything about the size: the shell owns it, because the shell is
         // what remembers it between launches.
         dataPlane.FontSizeStepRequested += (_, e) => FontSizeStepRequested?.Invoke(this, e);
+
+        // Fire-and-forget: this fires on the socket-accept thread, in the middle of the data plane's own
+        // handshake handling, and has no business making that wait on however long a replay takes. See
+        // ReplayAfterAttachAsync for what "replay" means and why racing the fresh page's own first frames
+        // is harmless.
+        dataPlane.SocketAttached += (_, _) => _ = ReplayAfterAttachAsync();
     }
 
     /// <summary>
@@ -224,7 +237,28 @@ public sealed class TerminalWorkspace : IAsyncDisposable
     }
 
     /// <summary>
-    /// Raised with the session id when a shell ends on its own.
+    /// A live session's flow-control window, or null when the id names no session this workspace still has
+    /// open.
+    /// </summary>
+    /// <remarks>
+    /// A test seam rather than something the shell has ever needed: nothing outside this assembly has a
+    /// reason to see a pump's credit window rather than what the transport does with it, but
+    /// <see cref="ReplayAfterAttachAsync"/>'s reset of that window on reattach is exactly the kind of thing
+    /// that is easy to get backwards, and worth asserting directly rather than only through its side
+    /// effects. Internal rather than public, reachable from the test assembly through the
+    /// <c>InternalsVisibleTo</c> this project already declares for it.
+    /// </remarks>
+    internal CreditWindow? CreditsFor(uint sessionId)
+    {
+        lock (sessionGate)
+        {
+            return sessions.TryGetValue(sessionId, out var session) ? session.Pump.Credits : null;
+        }
+    }
+
+    /// <summary>
+    /// Raised with the session id once a session is over — its shell having ended on its own, or a
+    /// deliberate close having fully drained.
     /// </summary>
     /// <remarks>
     /// <para>
@@ -235,13 +269,23 @@ public sealed class TerminalWorkspace : IAsyncDisposable
     /// the half of the interface Avalonia draws.
     /// </para>
     /// <para>
-    /// <b>Raised on whatever thread the pump finished on</b>, which is a thread-pool thread. A handler that
-    /// touches an observable collection has to marshal; this type has no toolkit to do it with, which is
-    /// exactly why it does not try.
+    /// <b>Raised only after the session's run task has completed, and that ordering is load-bearing.</b>
+    /// It used to fire from inside the run's own finally block, where the task is by definition not yet
+    /// complete — so a handler reading <see cref="LiveSessionCount"/> still counted the session that had
+    /// just ended, which is how the phone's foreground notification went on saying "1 shell connected"
+    /// over nothing. See <see cref="AnnounceEndedAsync"/>. Raised on a thread-pool continuation, or on the
+    /// closer's own thread; a handler that touches an observable collection has to marshal either way.
     /// </para>
     /// <para>
-    /// Not raised by <see cref="CloseSessionAsync"/>. That path already has a caller who knows the session is
-    /// going, and telling it what it just asked for is how a tab close turns into a second tab close.
+    /// <b>Raised by <see cref="CloseSessionAsync"/> too, which reverses a recorded decision.</b> The old
+    /// reasoning — the caller asked, so telling it is an echo — assumed every subscriber was the caller.
+    /// The phone's keep-alive is not: it hears this event to reconcile a notification with reality, and a
+    /// close that announced nothing left that notification claiming a shell that was gone. Every subscriber
+    /// treats the event as "reconcile" rather than "act" — a tab is marked dead if it is still there and
+    /// skipped if it is not — so a second announcement for a session that already announced its own end
+    /// (closing the tab of a shell that exited earlier) is deliberate and harmless. Shutdown is the one
+    /// close that stays silent: <see cref="DisposeAsync"/> is tearing the subscribers down with the
+    /// sessions, and news nobody is left to hear is not news.
     /// </para>
     /// </remarks>
     public event EventHandler<TerminalSessionEndedEventArgs>? SessionEnded;
@@ -254,6 +298,24 @@ public sealed class TerminalWorkspace : IAsyncDisposable
     /// </remarks>
     public event EventHandler<TerminalFontSizeStepEventArgs>? FontSizeStepRequested;
 
+    /// <summary>
+    /// Raised once a (re)attached renderer has been sent everything this workspace owns for it.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// The workspace's own share of "put the page back the way it was" is the sessions — each live one gets
+    /// its <c>SessionOpened</c> frame again, done by the time this fires. What is left is what the workspace
+    /// has no business owning: the font size and which tab is selected are both remembered by the shell, not
+    /// by a terminal, so this is the seam the shell uses to re-push them. See
+    /// <c>MainWindowViewModel</c>'s subscription for the other half.
+    /// </para>
+    /// <para>
+    /// Raised on the socket-accept thread, same as <see cref="TerminalDataPlane.SocketAttached"/> that
+    /// triggers it — a handler that touches a view model has to marshal.
+    /// </para>
+    /// </remarks>
+    public event EventHandler? RendererReattached;
+
     /// <summary>Starts the loopback listener.</summary>
     public void Start() => server = dataPlane.RunAsync(lifetime.Token);
 
@@ -352,6 +414,10 @@ public sealed class TerminalWorkspace : IAsyncDisposable
             sessions[sessionId] = new LiveSession(connection, pump, run);
         }
 
+        // The announcement's own continuation — see AnnounceEndedAsync. Started after the entry is stored,
+        // so the containment check inside it can never run against a dictionary the session had not reached.
+        _ = AnnounceEndedAsync(sessionId, run);
+
         return sessionId;
     }
 
@@ -523,6 +589,14 @@ public sealed class TerminalWorkspace : IAsyncDisposable
         {
             // Expected on the ordinary path: disposing the pump cancels its run.
         }
+
+        // After the drain, so a handler reading LiveSessionCount sees this session already gone — the
+        // event's own remark carries why a deliberate close is announced at all, and why shutdown is not:
+        // DisposeAsync sets the flag before its closing loop, and is dismantling every subscriber anyway.
+        if (Volatile.Read(ref disposed) == 0)
+        {
+            SessionEnded?.Invoke(this, new TerminalSessionEndedEventArgs(sessionId));
+        }
     }
 
     /// <inheritdoc />
@@ -563,6 +637,71 @@ public sealed class TerminalWorkspace : IAsyncDisposable
         lifetime.Dispose();
     }
 
+    /// <summary>
+    /// Rebuilds a freshly (re)attached page's idea of what is running, then tells the shell to rebuild its
+    /// own.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// Runs on the socket-accept thread that raised <see cref="TerminalDataPlane.SocketAttached"/> — the
+    /// constructor wires it up fire-and-forget for exactly that reason, so this method owns its own error
+    /// handling rather than leaving an unobserved exception for nobody to see.
+    /// </para>
+    /// <para>
+    /// Every live session — one whose <c>Run</c> has not completed — gets two things. Its credit window is
+    /// reset, because whatever was outstanding was reserved against bytes sent to a page that is now gone;
+    /// the acknowledgement that would return that credit died with it, and without this reset the session
+    /// would stall the moment 256 KiB of history had accumulated. And it gets its <c>SessionOpened</c> frame
+    /// again, marked with <see cref="ReplayMarker"/> so the page can tell a reattach from a session that is
+    /// genuinely new — the same frame a page that survived the socket drop already has a pane for, and one a
+    /// reloaded page does not.
+    /// </para>
+    /// <para>
+    /// A session whose shell has already ended gets nothing here. Its scrollback lived only in the page that
+    /// is gone, and sending a frame that implied otherwise would be exactly the kind of dishonesty this
+    /// fix is supposed to remove, not add. The tab strip still shows that session ended; nothing about this
+    /// method changes what <see cref="LiveSessionCount"/> or <see cref="IsSessionLive"/> report.
+    /// </para>
+    /// <para>
+    /// This can race the fresh page's own first frames — an early resize, an acknowledgement for output it
+    /// already had. That is harmless: every frame in both directions names its session, delivery order
+    /// within a session is preserved by both xterm and the socket, and a frame for a pane the page has not
+    /// created yet is simply dropped, the same as any frame for a session it does not know — see
+    /// <c>terminal.js</c>'s <c>handleFrame</c>.
+    /// </para>
+    /// </remarks>
+    private async Task ReplayAfterAttachAsync()
+    {
+        KeyValuePair<uint, LiveSession>[] live;
+
+        lock (sessionGate)
+        {
+            live = [.. sessions.Where(entry => !entry.Value.Run.IsCompleted)];
+        }
+
+        try
+        {
+            foreach (var (sessionId, session) in live)
+            {
+                session.Pump.Credits.Reset();
+
+                await dataPlane
+                    .SendAsync(
+                        TerminalFrame.Create((byte)TerminalServerOpcode.SessionOpened, sessionId, ReplayMarker),
+                        CancellationToken.None)
+                    .ConfigureAwait(false);
+            }
+
+            RendererReattached?.Invoke(this, EventArgs.Empty);
+        }
+        catch (Exception exception) when (exception is not OutOfMemoryException)
+        {
+            // Best-effort, same as every other fire-and-forget path here: a page that dies again mid-replay
+            // leaves nothing worse than the problem this method exists to fix, and there is no caller on
+            // this thread left to hand a failure to.
+        }
+    }
+
     private async Task RunSessionAsync(uint sessionId, TerminalSessionPump pump)
     {
         try
@@ -578,21 +717,51 @@ public sealed class TerminalWorkspace : IAsyncDisposable
             // the pump unwinding, which is why CloseSessionAsync needs no call of its own — and why this
             // must not do any work: it is running on a thread-pool thread inside DisposeAsync's loop when
             // the application is closing.
+            //
+            // SessionEnded is deliberately NOT raised from here, and it used to be — see
+            // AnnounceEndedAsync for what was wrong with that.
             ConnectionLog?.Closed(sessionId, clock.GetUtcNow());
+        }
+    }
 
-            // Only when the session is still one this workspace knows about. CloseSessionAsync removes the
-            // entry before it disposes the pump, so a tab the user closed does not come back as news.
-            bool announce;
+    /// <summary>Announces a session's end once its run task has actually completed.</summary>
+    /// <remarks>
+    /// <para>
+    /// A continuation rather than a line in <see cref="RunSessionAsync"/>'s finally, and the difference is
+    /// what a handler sees. Inside that finally the run task is not yet complete — a finally is part of the
+    /// task — so <see cref="LiveSessionCount"/>, which counts incomplete runs, still included the session
+    /// that had just ended. The phone's keep-alive answers this event by reading exactly that count, and
+    /// reconciled its foreground notification to "1 shell connected" over a shell that was gone, with
+    /// nothing left to fire afterwards and correct it. By the time an await on the run resumes, the task is
+    /// complete and the count is honest.
+    /// </para>
+    /// <para>
+    /// The containment check keeps the deliberate paths out of this route: <see cref="CloseSessionAsync"/>
+    /// removes the entry before it disposes the pump, and makes its own announcement after its own drain.
+    /// </para>
+    /// </remarks>
+    private async Task AnnounceEndedAsync(uint sessionId, Task run)
+    {
+        try
+        {
+            await run.ConfigureAwait(false);
+        }
+        catch (Exception exception) when (exception is not OutOfMemoryException)
+        {
+            // The run's faults belong to whoever drains it — CloseSessionAsync, on the deliberate path.
+            // This continuation cares only that the run is over, however it got there.
+        }
 
-            lock (sessionGate)
-            {
-                announce = sessions.ContainsKey(sessionId);
-            }
+        bool announce;
 
-            if (announce)
-            {
-                SessionEnded?.Invoke(this, new TerminalSessionEndedEventArgs(sessionId));
-            }
+        lock (sessionGate)
+        {
+            announce = sessions.ContainsKey(sessionId);
+        }
+
+        if (announce)
+        {
+            SessionEnded?.Invoke(this, new TerminalSessionEndedEventArgs(sessionId));
         }
     }
 

tests/DodoSSH.Client.App.Layout.Tests/ButtonCaptionTests.cs

@@ -1,13 +1,14 @@
 using System.Globalization;
 using Avalonia;
 using Avalonia.Controls;
+using Avalonia.Controls.Presenters;
 using Avalonia.Layout;
 using Avalonia.VisualTree;
 
 namespace DodoSSH.Client.App.Layout.Tests;
 
 /// <summary>
-/// The three button shapes centre their caption inside a button taller than the caption.
+/// The button shapes that centre their caption do, and the two that deliberately do not still fill.
 /// </summary>
 /// <remarks>
 /// <para>
@@ -41,6 +42,15 @@ namespace DodoSSH.Client.App.Layout.Tests;
 /// action — so an absolute expectation would be a font metric written down in a test file, and it would move
 /// the day the face does. "Centred" survives both.
 /// </para>
+/// <para>
+/// The five shapes beyond the original three were swept in afterwards, and none of them was misbehaving
+/// when it was: every one is content-sized everywhere it is used today, so <c>Stretch</c> and <c>Center</c>
+/// agreed and the change moved nothing — 113 buttons across 29 screens measured byte-identical before and
+/// after. What the sweep buys is that the day any of them is given a height, it is already right. That is
+/// also why <see cref="AStretchingShapeStillFillsItsButton"/> matters more than it looks: the same
+/// reasoning applied to <c>flat</c> or <c>cat</c> would break a pill and a strip that are currently
+/// correct.
+/// </para>
 /// </remarks>
 public sealed class ButtonCaptionTests
 {
@@ -61,6 +71,11 @@ public sealed class ButtonCaptionTests
     [InlineData("ghost")]
     [InlineData("accent")]
     [InlineData("danger")]
+    [InlineData("navuser")]
+    [InlineData("poprow")]
+    [InlineData("panechip")]
+    [InlineData("chiptoggle")]
+    [InlineData("choice")]
     public async Task ACaptionIsCentredInAButtonTallerThanItself(string shape)
     {
         await MeasureAsync(
@@ -96,6 +111,72 @@ public sealed class ButtonCaptionTests
                 "the caption should sit high, which is the defect this suite was written for"));
     }
 
+    /// <summary>
+    /// <c>flat</c> and <c>cat</c> are excluded from the rule above, and must stay excluded.
+    /// </summary>
+    /// <remarks>
+    /// <para>
+    /// Both stretch their content on purpose, and both would be silently broken by a later pass that
+    /// "finished" the sweep the rest of these classes belong to — which is exactly why this is a test and
+    /// not a comment.
+    /// </para>
+    /// <para>
+    /// <c>flat</c> carries the titlebar's search pill, a <c>Border.searchpill</c> with no height of its own
+    /// that is meant to fill all 35 pixels of the button; the usage states
+    /// <c>HorizontalContentAlignment="Stretch"</c> and relies on the vertical default matching it. Centring
+    /// from the style would shrink that pill to its caption's line box inside a button twice as tall.
+    /// <c>cat</c> carries the keychain rail's accent strip, a <c>Border.rowmark</c> whose style sets
+    /// <c>Width="2"</c> and no height at all — "at full row height", says the rule's own remark — so its
+    /// height is the stretch and nothing else.
+    /// </para>
+    /// <para>
+    /// Asserted as "the content fills the button", not as "the caption is off-centre": what these two need
+    /// is the fill, and a test phrased the other way would still pass if the fill broke in some new way.
+    /// </para>
+    /// </remarks>
+    [Theory]
+    [InlineData("flat")]
+    [InlineData("cat")]
+    public async Task AStretchingShapeStillFillsItsButton(string shape)
+    {
+        await LayoutHarness.OnTheUiThreadAsync(
+            () =>
+            {
+                // A bare Border is what both of them actually hold: no height, sized only by its parent.
+                var fill = new Border();
+                var button = new Button { Content = fill, Height = FixedHeight };
+                button.Classes.Add(shape);
+
+                var window = LayoutHarness.HostAtMinimumSize(
+                    button, LayoutHarness.MinimumWidth, LayoutHarness.MinimumHeight);
+
+                try
+                {
+                    // The slot read off the presenter rather than recomputed from the button's Padding:
+                    // these shapes differ in whether their presenter also draws a border, and a hand-rolled
+                    // sum was two pixels out on Button.cat for exactly that reason.
+                    var presenter = button.GetVisualDescendants()
+                        .OfType<ContentPresenter>()
+                        .Single(p => string.Equals(p.Name, "PART_ContentPresenter", StringComparison.Ordinal));
+
+                    var slot = presenter.Bounds.Height
+                        - presenter.Padding.Top - presenter.Padding.Bottom
+                        - presenter.BorderThickness.Top - presenter.BorderThickness.Bottom;
+
+                    fill.Bounds.Height.ShouldBe(
+                        slot,
+                        Tolerance,
+                        $"Button.{shape} must stretch its content — the search pill and the rail's accent "
+                        + "strip have no height of their own");
+                }
+                finally
+                {
+                    window.Close();
+                }
+            },
+            Token);
+    }
+
     private static Task MeasureAsync(
         string shape, VerticalAlignment? alignment, Action<double, double> assert) =>
         LayoutHarness.OnTheUiThreadAsync(

tests/DodoSSH.Client.App.Tests/ShellFlowTests.cs

@@ -2,6 +2,7 @@ using System.Globalization;
 using DodoSSH.Client.Auth;
 using DodoSSH.Client.Domain;
 using DodoSSH.Client.Import;
+using DodoSSH.Client.ObjectStore;
 using DodoSSH.Client.Session;
 // FakeDeviceKeyStore is compiled into this assembly from a source link and keeps its original namespace;
 // see the csproj for why it is shared rather than reimplemented.
@@ -1253,6 +1254,68 @@ public sealed class ShellFlowTests : IAsyncLifetime
         shell.IsTerminalShowing.ShouldBeTrue();
     }
 
+    /// <remarks>
+    /// The status line is one line for the whole window, so a sentence about a session that has been closed is
+    /// a sentence the user reads over some other terminal.
+    /// </remarks>
+    [Fact]
+    public async Task ClosingATab_TakesItsOwnStatusLineWithIt()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        await using var renderer = await FakeRenderer.AttachAsync(workspace, Token);
+
+        await vault.ConnectCommand.ExecuteAsync(null);
+        vault.Status.ShouldContain("Connected", Case.Insensitive);
+
+        await shell.CloseTabCommand.ExecuteAsync(shell.Tabs[0]);
+
+        vault.Status.ShouldBeEmpty();
+    }
+
+    /// <remarks>
+    /// The other half, and the reason closing does not simply blank the bar: everything this application has
+    /// to say about a save, a sync or a refusal goes through the same line — see <c>StatusBar.axaml</c> — and
+    /// closing a terminal answers none of it.
+    /// </remarks>
+    [Fact]
+    public async Task ClosingATab_LeavesAStatusLineThatIsAboutSomethingElse()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        await using var renderer = await FakeRenderer.AttachAsync(workspace, Token);
+
+        await vault.ConnectCommand.ExecuteAsync(null);
+        vault.Status = "The keychain could not be saved.";
+
+        await shell.CloseTabCommand.ExecuteAsync(shell.Tabs[0]);
+
+        vault.Status.ShouldBe("The keychain could not be saved.");
+    }
+
+    /// <remarks>
+    /// Two sessions to the same host, which is the case a label could not tell apart: the line belongs to the
+    /// attempt that wrote it, so closing the other one leaves it alone.
+    /// </remarks>
+    [Fact]
+    public async Task ClosingATab_LeavesTheLineAnotherTabWrote()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        await using var renderer = await FakeRenderer.AttachAsync(workspace, Token);
+
+        await vault.ConnectCommand.ExecuteAsync(null);
+        var first = shell.Tabs[0];
+
+        // The second connection's own line is what the bar holds now.
+        await vault.ConnectCommand.ExecuteAsync(null);
+        var reported = vault.Status;
+
+        await shell.CloseTabCommand.ExecuteAsync(first);
+
+        vault.Status.ShouldBe(reported);
+    }
+
     // ---- Connecting, while it is still happening ----
     //
     // A handshake is a network round trip and no longer holds the vault while it runs, so there is a stretch
@@ -8059,6 +8122,221 @@ public sealed class ShellFlowTests : IAsyncLifetime
         shell.Transfers.RemoteEntries.Select(entry => entry.Name).ShouldBe(["notes.txt"]);
     }
 
+    /// <remarks>
+    /// The phone's Files-screen chip row, proven at the view model rather than through Avalonia: a pin
+    /// saved on the host before this screen ever connects to it is read straight off the row's own
+    /// <c>HostSecret.PinnedPaths</c> at the moment <c>MarkHostConnected</c> runs, which is what
+    /// <see cref="TransfersViewModel.ConnectedPinnedPaths"/>'s own remark promises rather than a live follow
+    /// of the vault.
+    /// </remarks>
+    [Fact]
+    public async Task ConnectingATransfersHostWithPins_PopulatesConnectedPinnedPaths()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        vault.EditSelectedHostCommand.Execute(null);
+        vault.EditorNewPin = "/var/www/app";
+        vault.AddEditorPinCommand.Execute(null);
+        await vault.SaveHostCommand.ExecuteAsync(null);
+
+        shell.Transfers.Attach(vault, knownHosts);
+        shell.Transfers.SelectedHost = shell.Transfers.Hosts[0];
+
+        await shell.Transfers.ConnectCommand.ExecuteAsync(null);
+
+        shell.Transfers.IsConnected.ShouldBeTrue(shell.Transfers.Status);
+        shell.Transfers.ConnectedPinnedPaths.ShouldBe(["/var/www/app"]);
+        shell.Transfers.HasConnectedPins.ShouldBeTrue();
+    }
+
+    /// <remarks>
+    /// The other half of <see cref="ConnectingATransfersHostWithPins_PopulatesConnectedPinnedPaths"/>: the
+    /// chip row has to go with the connection it belongs to, or a later connect to a host with no pins would
+    /// show the previous host's.
+    /// </remarks>
+    [Fact]
+    public async Task DisconnectingTheTransfersScreen_ClearsConnectedPinnedPaths()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        vault.EditSelectedHostCommand.Execute(null);
+        vault.EditorNewPin = "/var/www/app";
+        vault.AddEditorPinCommand.Execute(null);
+        await vault.SaveHostCommand.ExecuteAsync(null);
+
+        shell.Transfers.Attach(vault, knownHosts);
+        shell.Transfers.SelectedHost = shell.Transfers.Hosts[0];
+
+        await shell.Transfers.ConnectCommand.ExecuteAsync(null);
+        shell.Transfers.HasConnectedPins.ShouldBeTrue();
+
+        await shell.Transfers.DisconnectCommand.ExecuteAsync(null);
+
+        shell.Transfers.ConnectedPinnedPaths.ShouldBeEmpty();
+        shell.Transfers.HasConnectedPins.ShouldBeFalse();
+    }
+
+    /// <remarks>
+    /// The phone's foreground-service question, proven at the view model rather than through Android: a
+    /// connect that opens an SFTP session is exactly the transition <c>SessionKeepAlive</c> needs to hear
+    /// about even when no transfer ever moves — see <see cref="TransfersViewModel.ActivityChanged"/>'s own
+    /// remark for why the queue's own raise, in <c>OnTransferChanged</c>, cannot cover a connect that never
+    /// touches <c>Transfers</c> at all.
+    /// </remarks>
+    [Fact]
+    public async Task ConnectingATransfersHost_RaisesActivityChangedAndTurnsOnHasLiveFileSession()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        shell.Transfers.Attach(vault, knownHosts);
+        shell.Transfers.SelectedHost = shell.Transfers.Hosts[0];
+
+        var raised = 0;
+        shell.Transfers.ActivityChanged += (_, _) => raised++;
+
+        await shell.Transfers.ConnectCommand.ExecuteAsync(null);
+
+        shell.Transfers.IsConnected.ShouldBeTrue(shell.Transfers.Status);
+        shell.Transfers.HasLiveFileSession.ShouldBeTrue();
+        raised.ShouldBeGreaterThan(0);
+    }
+
+    /// <remarks>
+    /// The other half: a disconnect is as much a transition the service must hear about as a connect is,
+    /// because it is the moment the connection <see cref="TransfersViewModel.HasLiveFileSession"/> promised
+    /// was open stops being true — and the foreground service would otherwise keep the process alive over a
+    /// session that has already closed.
+    /// </remarks>
+    [Fact]
+    public async Task DisconnectingTheTransfersScreen_RaisesActivityChangedAndTurnsOffHasLiveFileSession()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        shell.Transfers.Attach(vault, knownHosts);
+        shell.Transfers.SelectedHost = shell.Transfers.Hosts[0];
+
+        await shell.Transfers.ConnectCommand.ExecuteAsync(null);
+        shell.Transfers.HasLiveFileSession.ShouldBeTrue();
+
+        var raised = 0;
+        shell.Transfers.ActivityChanged += (_, _) => raised++;
+
+        await shell.Transfers.DisconnectCommand.ExecuteAsync(null);
+
+        shell.Transfers.HasLiveFileSession.ShouldBeFalse();
+        raised.ShouldBeGreaterThan(0);
+    }
+
+    /// <remarks>
+    /// A bucket is an <c>IRemoteFileStore</c> with no <c>HostSecret</c> underneath it, so there is no
+    /// <c>PinnedPaths</c> to read at all — see <see cref="TransfersViewModel.OpenBucketAsync"/>'s own remark.
+    /// The bucket here is created through the same keychain route
+    /// <see cref="TheS3ScreenWithNoBuckets_SaysWhereOneIsMadeAndGoesThere"/> exercises, and
+    /// <see cref="FakeObjectStoreFactory"/> stands in for the network the way <see cref="FakeSshConnectionFactory"/>
+    /// already does for SFTP.
+    /// </remarks>
+    [Fact]
+    public async Task ConnectingABucket_LeavesConnectedPinnedPathsEmpty()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        shell.Transfers.Attach(vault, knownHosts, buckets: new FakeObjectStoreFactory());
+
+        vault.NewObjectStoreCommand.Execute(null);
+        vault.BucketEditorLabel = "Backups";
+        vault.BucketEditorBucket = "backups";
+        vault.BucketEditorAccessKeyId = "AKIAEXAMPLE";
+        vault.BucketEditorSecretAccessKey = "a-secret-access-key";
+        vault.BucketEditorRegion = "eu-west-1";
+        await vault.SaveObjectStoreCommand.ExecuteAsync(null);
+
+        // Remote is what ConnectAsync branches on, and Attach's RefreshHosts has already auto-selected the
+        // host ReadyToConnectAsync left in the picker — without this line the command below dialled that
+        // host, and every assertion here passed only because that host happens to have no pins either. The
+        // ConnectedTo check is the proof the bucket path was actually taken.
+        shell.Transfers.Remote = RemoteKind.Bucket;
+        shell.Transfers.SelectedBucket = shell.Transfers.Buckets[0];
+
+        await shell.Transfers.ConnectCommand.ExecuteAsync(null);
+
+        shell.Transfers.ConnectedTo.ShouldBe("s3://backups");
+        shell.Transfers.IsConnected.ShouldBeTrue(shell.Transfers.Status);
+        shell.Transfers.ConnectedPinnedPaths.ShouldBeEmpty();
+        shell.Transfers.HasConnectedPins.ShouldBeFalse();
+    }
+
+    /// <remarks>
+    /// A bucket is HTTP, per-request, with nothing open that a dying process would lose — see
+    /// <see cref="TransfersViewModel.HasLiveFileSession"/>'s own remark. <c>IsConnected</c> alone would have
+    /// answered this wrongly, which is exactly why the flag reads <c>ConnectedCipher</c> as well: nothing
+    /// underneath a bucket ever sets it.
+    /// </remarks>
+    [Fact]
+    public async Task ConnectingABucket_LeavesHasLiveFileSessionOff()
+    {
+        var vault = await ReadyToConnectAsync();
+
+        shell.Transfers.Attach(vault, knownHosts, buckets: new FakeObjectStoreFactory());
+
+        vault.NewObjectStoreCommand.Execute(null);
+        vault.BucketEditorLabel = "Backups";
+        vault.BucketEditorBucket = "backups";
+        vault.BucketEditorAccessKeyId = "AKIAEXAMPLE";
+        vault.BucketEditorSecretAccessKey = "a-secret-access-key";
+        vault.BucketEditorRegion = "eu-west-1";
+        await vault.SaveObjectStoreCommand.ExecuteAsync(null);
+
+        // ReadyToConnectAsync already left a host in the picker, and Attach's own RefreshHosts auto-selects
+        // it — so without this the CONNECT command below would dial that host rather than open the bucket,
+        // and a host with no pins would make ConnectedPinnedPathsEmpty-style assertions pass for the wrong
+        // reason. Remote is what ConnectAsync actually branches on.
+        shell.Transfers.Remote = RemoteKind.Bucket;
+        shell.Transfers.SelectedBucket = shell.Transfers.Buckets[0];
+
+        await shell.Transfers.ConnectCommand.ExecuteAsync(null);
+
+        shell.Transfers.ConnectedTo.ShouldBe("s3://backups", "proof this opened the bucket rather than the host");
+        shell.Transfers.IsConnected.ShouldBeTrue(shell.Transfers.Status);
+        shell.Transfers.HasLiveFileSession.ShouldBeFalse();
+    }
+
+    /// <summary>A bucket that opens and lists as empty, so a bucket connect can be proven with no network.</summary>
+    private sealed class FakeObjectStoreFactory : IObjectStoreFactory
+    {
+        public IRemoteFileStore Open(ObjectStoreSecret store) => new FakeBucketStore();
+    }
+
+    /// <summary>The minimum <see cref="IRemoteFileStore"/> a bucket connect touches: home, then a listing.</summary>
+    private sealed class FakeBucketStore : IRemoteFileStore
+    {
+        public bool IsConnected => true;
+
+        public string HomeDirectory => "/";
+
+        public Task<IReadOnlyList<SftpEntry>> ListAsync(string path, CancellationToken cancellationToken) =>
+            Task.FromResult<IReadOnlyList<SftpEntry>>([]);
+
+        public Task<SftpEntry?> StatAsync(string path, CancellationToken cancellationToken) =>
+            Task.FromResult<SftpEntry?>(null);
+
+        public Task<Stream> OpenReadAsync(string path, long offset, CancellationToken cancellationToken) =>
+            throw new NotSupportedException("Not exercised by proving a bucket connect leaves no pins.");
+
+        public Task<Stream> OpenWriteAsync(string path, long offset, CancellationToken cancellationToken) =>
+            throw new NotSupportedException("Not exercised by proving a bucket connect leaves no pins.");
+
+        public Task CreateDirectoryAsync(string path, CancellationToken cancellationToken) =>
+            throw new NotSupportedException("Not exercised by proving a bucket connect leaves no pins.");
+
+        public Task DeleteAsync(string path, CancellationToken cancellationToken) =>
+            throw new NotSupportedException("Not exercised by proving a bucket connect leaves no pins.");
+
+        public Task RenameAsync(string fromPath, string toPath, CancellationToken cancellationToken) =>
+            throw new NotSupportedException("Not exercised by proving a bucket connect leaves no pins.");
+
+        public ValueTask DisposeAsync() => ValueTask.CompletedTask;
+    }
+
     /// <remarks>
     /// <para>
     /// The picker that replaced the desktop's connect bar, and the four things that put it away again. It

tests/DodoSSH.Client.Terminal.Tests/FakeShellSession.cs

@@ -8,6 +8,8 @@ internal sealed class FakeShellSession : ISshShellSession
     private readonly List<byte> written = [];
     private readonly Lock gate = new();
 
+    private readonly bool blockReads;
+
     private long remaining;
     private byte pattern;
 
@@ -16,7 +18,19 @@ internal sealed class FakeShellSession : ISshShellSession
     /// endless producer, which is what a runaway remote process looks like — those sessions are ended
     /// by disposing the pump rather than by running out of data.
     /// </param>
-    internal FakeShellSession(long bytesToProduce = 0) => remaining = bytesToProduce;
+    /// <param name="blockReads">
+    /// True for a shell that is open and live but has nothing to say — an idle prompt, rather than either
+    /// end of the "produces bytes" and "hit end of stream" spectrum <paramref name="bytesToProduce"/>
+    /// covers. <see cref="ReadAsync"/> then blocks until cancelled, which is what a real idle SSH channel's
+    /// read does. Exists for tests that need a session whose <c>Run</c> stays live without a background
+    /// read loop racing the test for control of the pump's credit window — see the reattach tests in
+    /// <c>TerminalWorkspaceTests</c>.
+    /// </param>
+    internal FakeShellSession(long bytesToProduce = 0, bool blockReads = false)
+    {
+        remaining = bytesToProduce;
+        this.blockReads = blockReads;
+    }
 
     /// <inheritdoc />
     public bool IsOpen { get; private set; } = true;
@@ -52,6 +66,13 @@ internal sealed class FakeShellSession : ISshShellSession
     {
         ReadCount++;
 
+        if (blockReads)
+        {
+            // Never completes on its own. The only way out is the same way a real blocked read ends: the
+            // token being cancelled, which is what disposing the pump does.
+            await Task.Delay(Timeout.InfiniteTimeSpan, cancellationToken).ConfigureAwait(false);
+        }
+
         await Task.Yield();
         cancellationToken.ThrowIfCancellationRequested();
 
@@ -109,7 +130,8 @@ internal sealed class FakeShellSession : ISshShellSession
 /// workspace is the layer that decides when a session is over, and that decision is what needs a
 /// connection whose shell can be made to end on cue.
 /// </remarks>
-internal sealed class FakeConnectionFactory(long bytesPerShell = long.MaxValue) : ISshConnectionFactory
+internal sealed class FakeConnectionFactory(long bytesPerShell = long.MaxValue, bool blockShellReads = false)
+    : ISshConnectionFactory
 {
     /// <summary>Connections handed out, in order.</summary>
     internal List<FakeConnection> Connections { get; } = [];
@@ -119,7 +141,7 @@ internal sealed class FakeConnectionFactory(long bytesPerShell = long.MaxValue)
         SshConnectionRequest request,
         CancellationToken cancellationToken)
     {
-        var connection = new FakeConnection(request, bytesPerShell);
+        var connection = new FakeConnection(request, bytesPerShell, blockShellReads);
         Connections.Add(connection);
 
         return Task.FromResult<ISshConnection>(connection);
@@ -127,7 +149,8 @@ internal sealed class FakeConnectionFactory(long bytesPerShell = long.MaxValue)
 }
 
 /// <summary>A connection that opens fake shells and records its own disposal.</summary>
-internal sealed class FakeConnection(SshConnectionRequest request, long bytesPerShell) : ISshConnection
+internal sealed class FakeConnection(SshConnectionRequest request, long bytesPerShell, bool blockShellReads = false)
+    : ISshConnection
 {
     /// <inheritdoc />
     public bool IsConnected { get; private set; } = true;
@@ -148,7 +171,7 @@ internal sealed class FakeConnection(SshConnectionRequest request, long bytesPer
     /// <inheritdoc />
     public Task<ISshShellSession> OpenShellAsync(TerminalSize size, CancellationToken cancellationToken)
     {
-        Shell = new FakeShellSession(bytesPerShell);
+        Shell = new FakeShellSession(bytesPerShell, blockShellReads);
 
         return Task.FromResult<ISshShellSession>(Shell);
     }

tests/DodoSSH.Client.Terminal.Tests/TerminalDataPlaneTests.cs

@@ -142,15 +142,96 @@ public sealed class TerminalDataPlaneTests : IAsyncDisposable
             await ConnectAsync(origin: "https://evil.example"));
     }
 
+    /// <remarks>
+    /// The truth this replaced: a second valid attach used to be a 409, on the theory that one renderer
+    /// lives for the whole process. Android's WebView does not honour that theory — its renderer process is
+    /// routinely killed and the page reloads with a fresh socket — so a second valid attach is now a
+    /// takeover. This asserts both halves: the newcomer gets the connection, and the displaced socket
+    /// actually goes rather than lingering as a phantom nothing is reading from.
+    /// </remarks>
     [Fact]
-    public async Task ASecondRenderer_IsRejected()
+    public async Task ASecondRenderer_TakesOver_AndTheFirstSocketIsDropped()
     {
         Start();
 
         using var first = await ConnectAsync();
         first.State.ShouldBe(WebSocketState.Open);
 
-        await Should.ThrowAsync<WebSocketException>(async () => await ConnectAsync());
+        using var second = await ConnectAsync();
+        second.State.ShouldBe(WebSocketState.Open);
+
+        // The first socket was aborted rather than closed gracefully — Abort skips the close handshake
+        // entirely, so there is no Close frame for this side to see coming. What a receive on it sees
+        // instead is the connection simply gone, which the client surfaces as an exception rather than as
+        // a state that quietly flips on its own; nothing here reads from the socket otherwise, so the
+        // state alone would not move.
+        var firstBuffer = new byte[16];
+        await Should.ThrowAsync<Exception>(async () =>
+            await first.ReceiveAsync(firstBuffer.AsMemory(), TestContext.Current.CancellationToken));
+
+        await using var session = new FakeShellSession(bytesToProduce: 64);
+        await using var pump = CreatePump(session);
+        plane.Register(SessionId, pump);
+
+        var run = pump.RunAsync(TestContext.Current.CancellationToken);
+
+        var opened = await ReceiveAsync(second);
+        opened.Opcode.ShouldBe((byte)TerminalServerOpcode.SessionOpened);
+
+        var output = await ReceiveAsync(second);
+        output.Opcode.ShouldBe((byte)TerminalServerOpcode.Output);
+        output.Payload.Length.ShouldBe(64);
+
+        await SendAsync(
+            second,
+            (byte)TerminalClientOpcode.Acknowledge,
+            TerminalFrame.CreateAcknowledgementPayload((uint)output.Payload.Length));
+
+        await run;
+    }
+
+    /// <remarks>
+    /// <para>
+    /// The other half of the takeover: a renderer process that dies without a close handshake — which is
+    /// what a killed Android WebView actually does, no FIN, nothing — must not fault the send path. A
+    /// faulted send would propagate into <see cref="TerminalSessionPump"/>'s flush loop and freeze a live
+    /// session; see <see cref="TerminalDataPlane.SendAsync"/>'s remark for why. Disposing the client socket
+    /// abruptly, with no close handshake sent, is the closest this harness gets to that: the server-side
+    /// socket is left believing itself open until it actually tries to write to it.
+    /// </para>
+    /// <para>
+    /// Driven straight through <see cref="TerminalDataPlane.SendAsync"/> rather than through a pump, because
+    /// a pump adds nothing here — the point is entirely about the transport's own contract, and a session
+    /// layered on top would only leave it unclear whether a passing test proved the transport never threw or
+    /// merely that the frames never happened to need a live socket.
+    /// </para>
+    /// </remarks>
+    [Fact]
+    public async Task SendAsync_DoesNotThrow_WhenTheAttachedRendererDiedWithoutClosing_AndAFreshAttachStillReceives()
+    {
+        Start();
+
+        var first = await ConnectAsync();
+        first.State.ShouldBe(WebSocketState.Open);
+        first.Dispose();
+
+        // Whether this particular send lands on the OS's send buffer before the peer's absence is noticed,
+        // or fails immediately, is not the point — either way it must not throw.
+        await Should.NotThrowAsync(async () =>
+            await plane.SendAsync(
+                TerminalFrame.Create((byte)TerminalServerOpcode.Output, SessionId, "before"u8.ToArray()),
+                TestContext.Current.CancellationToken));
+
+        using var second = await ConnectAsync();
+
+        await Should.NotThrowAsync(async () =>
+            await plane.SendAsync(
+                TerminalFrame.Create((byte)TerminalServerOpcode.Output, SessionId, "after"u8.ToArray()),
+                TestContext.Current.CancellationToken));
+
+        var output = await ReceiveAsync(second);
+        output.Opcode.ShouldBe((byte)TerminalServerOpcode.Output);
+        Encoding.UTF8.GetString(output.Payload).ShouldBe("after");
     }
 
     // ---- Frames ----

tests/DodoSSH.Client.Terminal.Tests/TerminalWorkspaceTests.cs

@@ -1,3 +1,6 @@
+using System.Globalization;
+using System.Net.WebSockets;
+using System.Text;
 using DodoSSH.Client.Ssh;
 
 namespace DodoSSH.Client.Terminal.Tests;
@@ -214,12 +217,15 @@ public sealed class TerminalWorkspaceTests
     }
 
     /// <remarks>
-    /// The event the tab strip listens to, so a dot can go out the moment a shell exits rather than at the
-    /// next thing that happens to repaint. Raised only when the session ended on its own: a tab the user
-    /// closed has a caller who already knows, and telling it would turn one close into two.
+    /// The event the tab strip and the phone's keep-alive listen to, so a dot can go out — and a foreground
+    /// notification can come down — the moment a shell exits rather than at the next thing that happens to
+    /// repaint. The count captured inside the handler is the sharper half of this test: the announcement
+    /// used to fire from inside the run's own finally block, where the run task is not yet complete, so
+    /// <c>LiveSessionCount</c> read from the handler still said 1 — and the phone's notification went on
+    /// claiming a shell that was gone, with nothing left to fire and correct it.
     /// </remarks>
     [Fact]
-    public async Task ASessionEndingOnItsOwnIsAnnounced()
+    public async Task ASessionEndingOnItsOwnIsAnnounced_AfterTheCountStoppedIncludingIt()
     {
         // A shell with no output to give: its first read returns 0, which is a remote closing the channel,
         // so the pump finishes with nobody asking it to.
@@ -228,10 +234,12 @@ public sealed class TerminalWorkspaceTests
         await using var workspace = CreateWorkspace(connections);
 
         var ended = new List<uint>();
+        var liveAtAnnouncement = -1;
         workspace.SessionEnded += (_, e) =>
         {
             lock (ended)
             {
+                liveAtAnnouncement = workspace.LiveSessionCount;
                 ended.Add(e.SessionId);
             }
         };
@@ -246,18 +254,33 @@ public sealed class TerminalWorkspaceTests
                 return ended.Contains(sessionId);
             }
         });
+
+        lock (ended)
+        {
+            liveAtAnnouncement.ShouldBe(0, "the announcement must wait for the run to actually complete");
+        }
     }
 
-    /// <inheritdoc cref="ASessionEndingOnItsOwnIsAnnounced" />
+    /// <remarks>
+    /// The reversal of a recorded decision, and the event's own remark carries why: a close used to be
+    /// announced to nobody, on the theory that the caller already knew — but the phone's keep-alive is not
+    /// the caller, and a close it never heard about left the foreground notification claiming a shell that
+    /// was gone. Announced once, after the drain, so the count a handler reads is already honest.
+    /// </remarks>
     [Fact]
-    public async Task ClosingASessionIsNotAnnouncedBack()
+    public async Task ClosingASessionIsAnnounced_OnceItHasDrained()
     {
         var connections = new FakeConnectionFactory();
 
         await using var workspace = CreateWorkspace(connections);
 
         var announcements = 0;
-        workspace.SessionEnded += (_, _) => Interlocked.Increment(ref announcements);
+        var liveAtAnnouncement = -1;
+        workspace.SessionEnded += (_, _) =>
+        {
+            liveAtAnnouncement = workspace.LiveSessionCount;
+            Interlocked.Increment(ref announcements);
+        };
 
         var sessionId = await workspace.OpenSessionAsync(
             Request(), TerminalSize.Default, TestContext.Current.CancellationToken);
@@ -265,12 +288,15 @@ public sealed class TerminalWorkspaceTests
         await workspace.CloseSessionAsync(sessionId);
 
         Volatile.Read(ref announcements)
-            .ShouldBe(0, "a close the caller asked for is not news to report back to it");
+            .ShouldBe(1, "a close is news to the keep-alive even though it is an echo to the closer");
+        liveAtAnnouncement.ShouldBe(0, "announced after the drain, so the count already excludes it");
     }
 
     /// <remarks>
     /// Disposal is the other path that closes sessions, because it is process shutdown. Asserted so
-    /// that the SSH connections are known to be released rather than assumed to be.
+    /// that the SSH connections are known to be released rather than assumed to be — and that these closes,
+    /// unlike a deliberate one, are announced to nobody: shutdown is dismantling every subscriber along
+    /// with the sessions, and news nobody is left to hear is not news.
     /// </remarks>
     [Fact]
     public async Task DisposingTheWorkspaceClosesEverySession()
@@ -279,6 +305,9 @@ public sealed class TerminalWorkspaceTests
 
         var workspace = CreateWorkspace(connections);
 
+        var announcements = 0;
+        workspace.SessionEnded += (_, _) => Interlocked.Increment(ref announcements);
+
         await workspace.OpenSessionAsync(
             Request(), TerminalSize.Default, TestContext.Current.CancellationToken);
         await workspace.OpenSessionAsync(
@@ -289,6 +318,86 @@ public sealed class TerminalWorkspaceTests
         workspace.LiveSessionCount.ShouldBe(0);
         connections.Connections.Count.ShouldBe(2);
         connections.Connections.ShouldAllBe(connection => connection.IsDisposed);
+        Volatile.Read(ref announcements).ShouldBe(0, "shutdown closes are not announced");
+    }
+
+    // ---- Reattach ----
+
+    /// <remarks>
+    /// <para>
+    /// The scenario the whole fix exists for: a page that lost its socket — killed WebView renderer, or
+    /// simply a reload — reattaches, and the session that was already running has to come back rather than
+    /// sit there forever with its output going nowhere.
+    /// </para>
+    /// <para>
+    /// The session's shell blocks on every read rather than producing output, which is what an idle prompt
+    /// looks like and — for this test — is what keeps its <c>Run</c> live without a background read loop
+    /// competing with this test over the credit window's exact value.
+    /// </para>
+    /// <para>
+    /// Neither assertion below polls, deliberately. The "before" one does not need to: reserving credit is
+    /// a synchronous call, so it is true the instant it returns. The "after" one does not need to either,
+    /// for a subtler reason — <see cref="TerminalWorkspace.ReplayAfterAttachAsync"/> calls
+    /// <c>Credits.Reset()</c> and only then awaits sending the replay frame for that same session, with no
+    /// suspension between the two, so by the time this test has received that frame the reset has
+    /// necessarily already happened. A poll here would only have hidden a real ordering bug behind a
+    /// generous timeout instead of catching it.
+    /// </para>
+    /// </remarks>
+    [Fact]
+    public async Task ANewRenderer_ReplaysTheLiveSessionAndResetsItsCredits()
+    {
+        var connections = new FakeConnectionFactory(blockShellReads: true);
+
+        await using var workspace = CreateWorkspace(connections);
+        workspace.Start();
+
+        using var first = await ConnectRendererAsync(workspace);
+
+        var sessionId = await workspace.OpenSessionAsync(
+            Request(), TerminalSize.Default, TestContext.Current.CancellationToken);
+
+        // The session's own opening frame, sent as soon as the pump starts running. Not a replay, and not
+        // what this test is about — read and discarded so it cannot be confused for one below.
+        await ReceiveFrameAsync(first);
+
+        var credits = workspace.CreditsFor(sessionId).ShouldNotBeNull();
+        credits.TryReserve(4096);
+        credits.Outstanding.ShouldBeGreaterThanOrEqualTo(
+            4096, "the pump's own read loop may have reserved a buffer's worth on top of this");
+
+        using var second = await ConnectRendererAsync(workspace);
+
+        var replay = await ReceiveFrameAsync(second);
+        replay.Opcode.ShouldBe((byte)TerminalServerOpcode.SessionOpened);
+        replay.SessionId.ShouldBe(sessionId);
+        replay.Payload.ShouldBe(new byte[] { 1 }, "a replay is flagged so the page can tell it apart from a fresh open");
+
+        credits.Outstanding.ShouldBe(0, "the replay frame above cannot have been sent before the reset that precedes it");
+    }
+
+    /// <remarks>
+    /// The other half of a reattach: the workspace has replayed what it owns, and this is the seam the
+    /// shell uses to replay what it owns instead — the font size and the selected tab, neither of which a
+    /// terminal session knows anything about. <c>MainWindowViewModel</c>'s subscription is what actually
+    /// does that; this only asserts that the workspace hands it the chance to.
+    /// </remarks>
+    [Fact]
+    public async Task ANewRenderer_RaisesRendererReattached()
+    {
+        var connections = new FakeConnectionFactory();
+
+        await using var workspace = CreateWorkspace(connections);
+        workspace.Start();
+
+        using var first = await ConnectRendererAsync(workspace);
+
+        var reattachedCount = 0;
+        workspace.RendererReattached += (_, _) => Interlocked.Increment(ref reattachedCount);
+
+        using var second = await ConnectRendererAsync(workspace);
+
+        await WaitUntilAsync(() => Volatile.Read(ref reattachedCount) > 0);
     }
 
     // ---- Helpers ----
@@ -362,12 +471,76 @@ public sealed class TerminalWorkspaceTests
     private static InMemoryTerminalAssetProvider StubAssets() =>
         new(new Dictionary<string, TerminalAsset>(StringComparer.Ordinal)
         {
-            [TerminalDataPlane.PagePath] = new("text/html; charset=utf-8", "<!doctype html>"u8.ToArray()),
+            // The placeholders, not a token and URL already filled in — the reattach tests below have to
+            // connect a real renderer, and doing that by reading them back out of the served page is what
+            // proves the workspace serves a page a real renderer could actually attach with, rather than
+            // one that merely looks servable.
+            [TerminalDataPlane.PagePath] = new(
+                "text/html; charset=utf-8",
+                Encoding.UTF8.GetBytes(
+                    $"<html><body data-token=\"{TerminalDataPlane.TokenPlaceholder}\" "
+                    + $"data-socket=\"{TerminalDataPlane.SocketUrlPlaceholder}\"></body></html>")),
         });
 
     private static SshConnectionRequest Request() =>
         new("host.invalid", 22, "dodo", new SshPasswordCredential("irrelevant"));
 
+    /// <remarks>
+    /// Attaches the way the real page does: by fetching the served page, reading the token and socket URL
+    /// back out of it, and presenting them on the upgrade — rather than reaching into the workspace for a
+    /// token it does not expose. A shortcut here would prove only that a socket can be opened, not that the
+    /// workspace serves a page a renderer could actually attach with.
+    /// </remarks>
+    private static async Task<ClientWebSocket> ConnectRendererAsync(TerminalWorkspace workspace)
+    {
+        using var http = new HttpClient();
+        var page = await http.GetStringAsync(workspace.PageUrl, TestContext.Current.CancellationToken);
+
+        var token = ExtractAttribute(page, "data-token");
+        var socketUrl = ExtractAttribute(page, "data-socket");
+
+        var client = new ClientWebSocket();
+        client.Options.AddSubProtocol(TerminalDataPlane.SubProtocol);
+        client.Options.AddSubProtocol($"token.{token}");
+        client.Options.SetRequestHeader(
+            "Origin",
+            string.Create(CultureInfo.InvariantCulture, $"http://127.0.0.1:{workspace.PageUrl.Port}"));
+
+        try
+        {
+            await client.ConnectAsync(new Uri(socketUrl), TestContext.Current.CancellationToken);
+        }
+        catch
+        {
+            client.Dispose();
+            throw;
+        }
+
+        return client;
+    }
+
+    private static string ExtractAttribute(string html, string name)
+    {
+        var marker = $"{name}=\"";
+        var start = html.IndexOf(marker, StringComparison.Ordinal) + marker.Length;
+        var end = html.IndexOf('"', start);
+
+        return html[start..end];
+    }
+
+    private static async Task<(byte Opcode, uint SessionId, byte[] Payload)> ReceiveFrameAsync(
+        ClientWebSocket socket)
+    {
+        var buffer = new byte[64 * 1024];
+
+        var result = await socket.ReceiveAsync(buffer.AsMemory(), TestContext.Current.CancellationToken);
+
+        TerminalFrame.TryRead(buffer.AsSpan(0, result.Count), out var opcode, out var sessionId, out var payload)
+            .ShouldBeTrue();
+
+        return (opcode, sessionId, payload.ToArray());
+    }
+
     /// <remarks>
     /// Polled rather than awaited on a task, because the point is what an observer of the property
     /// sees: the pump ends on a thread of its own, and the count has to catch up without anyone