231 Commits

• aea: create XAppleArchive module
  Kabir Oberai · 2026-11-08 19:55:27 -05:00
  46230d8530
• anisette: fix crash in date formatter (#422)

  ## What does this PR do?
  
  Password login can terminate with an illegal-instruction crash on Linux,
  leaving the user logged out. Anisette client time was formatted with
  DateFormatter and an explicitly assigned Gregorian calendar. That
  formatting path reaches Foundation’s NSCalendar bridging code, where the
  reported crash occurs.
  
  AnisetteData now formats client time with Date.ISO8601FormatStyle,
  avoiding the DateFormatter calendar bridge. The header retains its UTC
  timestamp format with whole seconds and a trailing Z. New parameterized
  tests check the complete header dictionary for the Unix epoch, a year
  boundary, leap day, and a timestamp with fractional seconds.
  
  Fixes #184
  
  ## How was it tested?
  
  Ran `swift test` locally on macOS for the xtool AnisetteData change; it
  fails the same way on the base branch, so the failure predates this
  change (it fails the same way on upstream/main).
  `swift build` failed because an async call in a defer body prevented
  compilation.
  `swift test` failed during compilation with the same async-in-defer
  error, so the tests did not run.
  `repo_lint` ran without reported output.
  `make lint SWIFTLINT_FLAGS='--reporter=github-actions-logging' | tee
  swiftlint.log` ran with no lint diagnostics shown in the transcript.
  
  The tests in `Tests/XToolTests/AnisetteDataTests.swift` fail without the
  fix in the xtool AnisetteData change (checked by reverting the source
  changes).
  
  ## AI tool usage
  
  How much of the code in this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI in writing code. I reviewed the
  finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  AI was used for assistance.
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Client timestamps now use a consistent UTC ISO 8601 format with dashed
  dates, colon-separated times, and no fractional seconds. This applies to
  timestamps with or without fractional seconds in the source date.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  ---------
  
  Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>

  Matt Van Horn · 2026-11-06 00:17:14 -05:00
  4ec0e96264
• ds: fix profile creation conflicts (#430)

  ## What does this PR do?
  
  Fixes issues with missing devices and duplicate names when creating new
  profiles. Supersedes #427 and #428.
  
  Fixes #425, fixes #426
  
  ## How was it tested?
  
  - [x] Ran locally
  - [x] Confirmed that free accounts also support multiple profiles as
  long as the names are different.
  
  ## AI tool usage
  
  How much of the code in this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI in writing code. I reviewed the
  finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Profile creation now uses only enabled devices that support iOS apps.
  * Existing profiles with the exact generated profile name are removed
  before a replacement is created, including when multiple matches exist.
    * Profile handling now stops when the operation is canceled.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  Co-authored-by: Valentin Brosseau <c4software@gmail.com>

  Kabir Oberai · 2026-11-06 00:15:47 -05:00
  ac40282dbd
• utils: support forwarding Subprocess output (#413)

  ## What does this PR do?
  
  When spawning a subprocess,
  `currentStandardOutput`/`currentStandardError` bypass `d.out`. Add
  `.xtoolOutput()` and `.xtoolError()` which respect the dependency.
  
  ## How was it tested?
  
  Tested in a downstream PR to ensure that forwarding works
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **New Features**
  * Subprocess output can now be routed directly to XTool’s standard
  output or standard error.
  * Added a way to create a writable output descriptor that forwards data
  to a selected stream.
  * Asynchronous input reading is now available through a public
  interface, including configurable read sizes and explicit closing.
  
  * **Compatibility**
    * Existing input-reading behavior and parameters remain unchanged.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-31 01:36:50 -04:00
  bec5aa106a
• gsa: make more reqs ephemeral (#417)

  ## What does this PR do?
  
  I'm not sure if this is needed but, for good measure, updates remaining
  GSA requests to use an ephemeral client (followup from #247)
  
  ## How was it tested?
  
  - [x] Tested flows involving all of these requests
  
  ## AI tool usage
  
  How much of the code in this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI in writing code. I reviewed the
  finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Improvements**
  * Selected account and lookup requests now use temporary HTTP
  connections. Existing request details, response handling, and progress
  reporting remain supported, while other requests continue through the
  standard connection flow. This change applies across the app’s related
  account and lookup operations; no new user-facing controls or workflows
  are introduced.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-30 21:54:54 -04:00
  804cf3e1dd
• gsa: more SMS 2FA fixes (#416)

  ## What does this PR do?
  
  Applies the 2FA SMS login fixes from #405
  
  ## How was it tested?
  
  - [x] 2FA SMS arrives
  - [x] Can log in with SMS code
  
  ## AI tool usage
  
  How much of the code in this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI in writing code. I reviewed the
  finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Improvements**
  * SMS verification requests now use an updated phone verification format
  and numeric phone-number IDs.
  * Code validation also uses numeric phone-number IDs, keeping the
  identifier format consistent across SMS authentication steps. The
  verification request and code-validation request now follow the same
  updated format.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  Co-authored-by: Corne Steenhuis <cornesteenhuis@hotmail.com>

  Kabir Oberai · 2026-10-30 21:54:54 -04:00
  2186b72383
• gsa: GrandSlamMethod -> GrandSlamRequestContent (#415)

  ## What does this PR do?
  
  Renames GrandSlamMethod to GrandSlamRequestContent and adds support for
  json & PUT requests
  
  ## How was it tested?
  
  - [x] Still compiles
  - [x] `xtool auth`
  
  ## AI tool usage
  
  How much of the code in this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI in writing code. I reviewed the
  finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **New Features**
  * Requests can now send either JSON or XML bodies using POST or PUT. GET
  requests are sent without a body.
  * **Improvements**
  * Authentication and verification requests now use a consistent request
  format while retaining their existing behavior.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-30 21:54:54 -04:00
  90fe13fc9e
• signer: replace zsign with RCodesign (#389)

  ## What does this PR do?
  
  Replaces zsign with the
  [apple-codesign](https://github.com/indygreg/apple-platform-rs/tree/0ebbd2ef3b96d5adccc30b1cad4deea38cd735c4/apple-codesign)
  Rust crate. It's much more usable as a library (while zsign has bare
  prints all over the place + we had to fork it to maintain a public
  interface), it supports macOS-style bundles, and also has other features
  that we would've needed to further patch zsign for.
  
  ## How was it tested?
  
  - [x] Ran an ad-hoc signed app on the simulator (with a subsequent PR)
  - [x] Ran an app with App Groups on device, confirmed that the app group
  is present
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Updated Signing**
  * App signing now uses an updated signing engine and supports
  certificate-based and ad hoc signing.
  * Entitlement data, including data provided through entitlement files,
  is passed through during signing.
    * Deep signing is enabled in the development command.
  * **Validation**
  * Signing rejects invalid app or entitlement file paths and duplicate
  entitlement paths.
  * **Compatibility**
    * Signing contexts can be created without selecting a signer.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-30 02:47:23 -04:00
  64fe8e30b3
• ds: allow one app group on free teams (#284)

  ## What does this PR do?
  
  Resolves the FIXME in `DeveloperServicesCapability`: App Groups are no
  longer filtered out on free teams, and
  `DeveloperServicesAddAppOperation` assigns at most one group per app on
  a free team (the constraint the FIXME mentioned).
  
  Free teams can register and assign app groups through the Xcode
  DeveloperServices API (as AltStore/SideStore do), so a widget and its
  app can share `UserDefaults(suiteName:)` with a free Apple ID.
  
  ## How was it tested?
  
  Built on Linux from `main` (4d0c429). Installed a Flutter app with a
  WidgetKit extension on an iPhone (iOS 27.2) with a free Apple ID: the
  app gets `group.XTL-<team>.<id>` in its entitlements and the widget
  reads the app's data through it. `make lint` not run (no SwiftLint on
  this machine).
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
    * App Groups are now available to free teams.
  * Apps assigned to multiple App Groups are rejected with a clear error
  message.
  * Invalid app errors now include the app’s URL path, making it easier to
  identify which app could not be added.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  ---------
  
  Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
  Co-authored-by: Kabir Oberai <oberai.kabir@gmail.com>

  Cesar Andres Pereira Bernal · 2026-10-30 02:11:54 -04:00
  0c63e6979f
• ds: fix assigning multiple app groups (#403)

  ## The bug
  
  `DeveloperServicesAssignAppGroupsOperation` sends one
  `assignApplicationGroupToAppId` request per app group, concurrently.
  That endpoint replaces the App ID's groups with the ones in the request,
  so an app whose entitlements list two or more groups ends up with only
  one of them on the App ID: whichever request Apple handles last. The
  provisioning profile then carries one group while the signed binary
  claims two, and the device refuses the install with
  `ApplicationVerificationFailed`.
  
  Setting the groups by hand in the developer portal does not help,
  because the next `xtool dev build --sign` sends the per-group requests
  again.
  
  ## The fix
  
  - `DeveloperServicesAssignAppGroupRequest` takes `groupIDs:
  [DeveloperServicesAppGroup.ID]` and sends `applicationGroups` as an
  array. The endpoint accepts an array; it was being given a single string
  before.
  - `DeveloperServicesAssignAppGroupsOperation.perform()` still finds or
  creates each group concurrently, then sends a single assign request
  carrying every group ID, and returns the group IDs in the order the
  entitlements listed them, so the rewritten entitlement is stable between
  builds.
  
  The only caller, `DeveloperServicesAddAppOperation`, is unchanged.
  
  ## Related
  
  #324 would report the resulting mismatch between the entitlements and
  the profile as a clear error at signing time; this change removes the
  cause, so an app with several groups gets a profile that grants all of
  them.
  
  ## Checked
  
  Built on Linux with the Swift 6.4 toolchain and used to sign an iOS app
  whose main target has two app groups and a Share extension. The app's
  `embedded.mobileprovision` now carries both groups and the device
  installs the app. The same app signed with the unpatched build got one
  group in the profile and `ApplicationVerificationFailed` on install.
  
  🤖 Generated with [Claude Code](https://claude.com/claude-code)
  
  https://claude.ai/code/session_01UKzxgxaLwK7Ap5LMwj9MxH
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **New Features**
  * Apps can now be assigned multiple app groups in a single operation.
  The assignment replaces the app’s existing group associations with the
  selected groups, so any groups not included in the new assignment are no
  longer associated with the app.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  ---------
  
  Co-authored-by: David Reinecke <david.reinecke@gmail.com>
  Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
  Co-authored-by: Kabir Oberai <oberai.kabir@gmail.com>

  David Reinecke · 2026-10-30 05:38:25 +00:00
  bca6f59848
• xcode: track paths on XcodeProject (#388)

  ## What does this PR do?
  
  An XcodeProject/XcodeWorkspace now knows its own location on disk, which
  is important when we start using these types to build PIF.
  
  ## How was it tested?
  
  - [x] Used in future PRs to build various Xcode projects
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Improvements**
  * Projects and workspaces now retain their associated locations and save
  directly to those locations.
  * Generated project files preserve the configured project directory
  path, helping keep file references consistent when projects are written
  or reopened.
  * Projects and workspaces can be associated with a new location while
  retaining their existing contents.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-28 19:22:27 -04:00
  f1b455b4c5
• utils: better XToolCache location (#409)

  ## What does this PR do?
  
  Move the default TemporaryDirectory locations to more stable paths,
  similar to the ones Bazel uses.
  
  - on macOS we now default to `\(URL.cachesDirectory)/sh.xtool` (usually
  `~/Library/Caches/sh.xtool`)
  - on Linux we (continue to) default to
  `${XDG_CACHE_HOME:-~/.cache}/xtool`
  - we no longer respect `$TMPDIR`. One can still use `$XTL_TMPDIR` to
  override.
  
  ## How was it tested?
  
  - [x] Cache directory locations are as expected
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Temporary-directory selection now ignores empty `XTL_TMPDIR` values
  and no longer falls back to `TMPDIR`.
  * Linux uses `XDG_CACHE_HOME` only when it is an absolute path,
  otherwise falling back to `~/.cache`.
  * **Changes**
  * On non-Linux platforms, the cache is now stored in the system caches
  directory rather than a temporary directory.
  * **Documentation**
    * Updated the documented temporary-directory settings.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-28 19:22:26 -04:00
  c34e9241d5
• utils: tweaks (#387)

  ## What does this PR do?
  
  Small utils refactors to make subsequent PRs cleaner.
  
  - Minor tweaks to how we determine / handle the xtool build version
  - Factor out TemporaryDirectory cache location
  
  ## How was it tested?
  
  - [x] Cache directory logic is unchanged
  - [x] Checked `xtool --version`
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **New Features**
  * Version information now distinguishes clean builds from builds with
  uncommitted changes, displaying a `-dirty` suffix when applicable.
  * Nonblocking file-locking operations are now available to external
  consumers.
  * **Changes**
  * Version reporting uses `XTOOL_VERSION` when set; otherwise, it uses
  available Git tag or commit information.
  * Temporary files now use the tool’s cache directory, honoring
  `XTL_TMPDIR` and platform cache settings. On Linux, the location follows
  `XDG_CACHE_HOME` when it is an absolute path; otherwise, the user cache
  directory is used.
    * `TMPDIR` is no longer used to select the temporary-file location.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-28 19:22:26 -04:00
  2f2336cb5f
• cli: better IO (#375)

  ## What does this PR do?
  
  Adds `Dependency`s for console IO. Also fixes #268 by explicitly
  handling non-pipe input.
  
  ## How was it tested?
  
  Ran on macOS + Linux in Docker. I have a future PR that overrides the
  dependencies in certain cases, which also appears to work correctly.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  Did not scrutinize the tests too much tbh but they look reasonable. Easy
  enough to update/remove them in the future if needed.
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **New Features**
  * Added support for reading command-line input, including asynchronous
  prompts, line-based responses, and graceful handling of end-of-input.
    * Added coordinated cleanup when commands finish.
  * **Improvements**
  * Command messages across authentication, device management,
  installation, SDK, and other workflows now use consistent output
  handling; displayed messages and command behavior remain unchanged.
  * Improved handling of complete output when the system temporarily
  cannot accept more data.
  * **Tests**
  * Added coverage for command input and output, including stream
  behavior, cancellation, and shutdown.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-28 17:48:38 -04:00
  3e0c11d967
• gsa: update user agent (#406)

  ## What does this PR do?
  
  Fixes the latest source of HTTP 503s during Apple ID login
  
  ## How was it tested?
  
  - [x] `xtool auth login`
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Improvements**
  * Requests now use the user-agent value provided with device
  information, rather than a fixed value.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-25 19:54:57 -04:00
  70e1568fdf
• pack: link edited dependencies into the builder package (#390)

  ## What does this PR do?
  
  SwiftPM looks for edited dependencies in the root package's Packages
  directory. The builder package becomes the root when building, so
  dependencies edited with `swift package edit` were ignored and SwiftPM
  fetched them again instead.
  
  <!-- Focus on *what* the PR does in essence. The *how* should come
  second, if it's relevant. -->
  
  ## How was it tested?
  
  Without this change, the build of `xtool/.xtool-tmp` tries to clone all
  15 dependencies and fails.
  
  * With `Packages` linked into `.xtool-tmp`, it builds without fetching
  anything and produces `DemoTCA.app`
  * `Examples/DemoSimple`, which has no `Packages` directory, still builds
  * `swift test` passes
  
  <!-- e.g. Ran locally on macOS and added unit tests -->
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Build Improvements**
  * Builds now include access to the working directory’s `Packages` folder
  when it exists. If the folder is absent, builds proceed without it. This
  can make packages in that folder available during the build.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  Signed-off-by: Anish Pallati <i@anish.land>

  Anish Pallati · 2026-10-23 17:48:49 -04:00
  7f7d5b7e44
• sdk: accept directories for tarball overrides (#385)

  ## What does this PR do?
  
  XTL_TOOLSET_TARBALL and XTL_OAM_TARBALL can now point at an unpacked
  directory, so package managers can use tools they already provide
  without repacking them into tarballs.
  
  <!-- Focus on *what* the PR does in essence. The *how* should come
  second, if it's relevant. -->
  
  ## How was it tested?
  
  Tested on Linux with Swift 6.2.4: 
  * `xtool sdk build` with both overrides set to directories in the Nix
  store
  * `xtool dev` flags the SDK as out of date after an override path
  changes
  
  <!-- e.g. Ran locally on macOS and added unit tests -->
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **New Features**
  * SDK setup now accepts a directory as a package source, recursively
  copying its contents and adding owner write permission.
  * Archive files continue to be supported as package sources and are
  extracted as before.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  Signed-off-by: Anish Pallati <i@anish.land>

  Anish Pallati · 2026-10-23 18:24:11 +00:00
  f1eca85cf4
• entitlements: support Sign in with Apple (#377)

  ## What does this PR do?
  
  Hi! xtool didn't know about the `com.apple.developer.applesignin`
  entitlement, so every time I installed my app, the capability sync
  removed Sign in with Apple from the App ID. The new provisioning profile
  didn't have it, and the device refused the app ("A valid provisioning
  profile for this executable was not found").
  
  This maps the entitlement to the `APPLE_ID_AUTH` capability, as a
  primary App ID, so the capability stays and the profile includes it.
  
  ## How was it tested?
  
  I built xtool from this branch on macOS and installed my app (a SwiftPM
  app with `entitlementsPath` and the Sign in with Apple entitlement) on
  an iPhone with a paid team. The App ID kept the capability, the install
  worked, and native Sign in with Apple works in the app. Before the
  change the same install failed every time.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **New Features**
    * Added support for configuring the Apple Sign In entitlement.
  * Apple Sign In is now available as a supported entitlement type,
  including serialization and configuration support.
    * Configures Apple ID app consent as the primary app consent option.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Víctor Agudo Gonzalez · 2026-10-22 22:25:37 -04:00
  952a7ffd43
• gsa: use macOS Omnisette (#380)

  ## What does this PR do?
  
  Fixes #378, solving error -80046. Seems Apple now blocks
  Android-identifying clients from using gsa (at least the 2fa bits). Use
  Mac Anisette instead. cf
  https://github.com/nab138/iloader/commit/902597564315442f00ab299dcc758ee0e13b8fa8
  
  Alas, this does mean we have to retire xadi and switch back to
  Omnisette.
  
  ## How was it tested?
  
  - [x] Ran locally on macOS
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  <!-- coderabbit-ai-generated:v1 -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
    * Two-factor requests now use the device’s configured User-Agent.
  * **Updates**
  * Anisette requests now use the Omnisette provider and a new default
  server. The optional server URL override remains available.
  * The password-login warning now explains the virtual Mac and
  third-party service involved, clarifies that credentials are sent only
  to Apple, and recommends considering a secondary Apple ID.
  * **Behavior Changes**
    * The Android-specific ADI provider is no longer available.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-22 17:04:16 -04:00
  e3ce2c402e
• sdk: support env overrides for tarballs (#372)

  ## What does this PR do?
  
  When building the sdk, allows you to set
  `XTL_TOOLSET_TARBALL`/`XTL_OAM_TARBALL` to use local tarballs instead of
  fetching them remotely. Supersedes
  https://github.com/xtool-org/xtool/pull/319.
  
  ## How was it tested?
  
  Honestly didn't bother but I think it works. Fairly safe to land.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  - **New Features**
  - Added support for installing the Toolset and OpenAppleMacros from
  locally provided tarball files.
  - Existing HTTP download behavior remains available when local tarballs
  are not configured.
  
  - **Chores**
    - Improved handling of lengthy installation logic.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  Co-authored-by: Mathieu Post <mathieupost@gmail.com>

  Kabir Oberai · 2026-10-20 01:00:51 -04:00
  aa6ba75e7d
• sdk: update OpenAppleMacros to v1.4.1 (#371)

  ## What does this PR do?
  
  Updates OpenAppleMacros to v1.4.1. It now includes an info.json that
  lists supported modules, so we don't need to hardcode them.
  
  ## How was it tested?
  
  Built a project that uses `@State` on Linux
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  - **New Features**
  - Updated the DemoSimple example with a SwiftUI app window, interactive
  counter, and button-based incrementing.
  - The example now runs its demonstration action when the content view
  loads.
  
  - **Bug Fixes**
  - Improved SDK component downloads with clearer progress reporting and
  validation.
  - Improved plugin setup across supported platforms for more consistent
  availability.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-20 01:00:50 -04:00
  ef7ce5b30d
• gsa: fall back to a trusted-device code when SMS 2FA fails (#349)

  `secondaryAuth` can't finish logging in: GSA rejects POST
  /auth/verify/phone/put?mode=sms with HTTP 403 ("Could not connect to
  iCloud"). The JSON flavor of those endpoints (GET /auth, PUT
  /auth/verify/phone/) answers 401 as well.
  
  GsService2/validate still accepts a code generated on a trusted device
  (Settings > Sign-In & Security > Get Verification Code), so ask for one
  instead of failing. Tested on WSL (Ubuntu 26.04) with an iOS 27 device.
  
  ## What does this PR do?
  
  <!-- Focus on *what* the PR does in essence. The *how* should come
  second, if it's relevant. -->
  
  ## How was it tested?
  
  <!-- e.g. Ran locally on macOS and added unit tests -->
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Two-factor authentication can fall back to a trusted device when an
  SMS request returns HTTP 403.
  * If trusted-device authentication also returns HTTP 403, users receive
  instructions for manually generating a verification code, and code
  validation can continue. Other errors are still reported.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  ---------
  
  Co-authored-by: Kabir Oberai <oberai.kabir@gmail.com>

  javiertrejoyam · 2026-10-19 00:55:16 +00:00
  8c46156e27
• auth: clean up Anisette handling (#367)

  ## What does this PR do?
  
  Simplifies the Anisette handling code. Removes some unnecessary fields
  (including hard-coded Mac serial numbers, cf #348) from our http
  requests amongst other things.
  
  ## How was it tested?
  
  Tested on macOS:
  
  - [x] Can log in after resetting provisioning
  - [x] Can run `xtool ds teams`
  - [x] Can run `xtool ds certificates`
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Updates**
  * Anisette-backed requests now use device information provided with
  Anisette data, including an Android 14 device identity.
  * Developer service requests no longer include client-information and
  device-ID headers.
  * The Anisette command now displays fetched details in sorted order and
  uses the configured Anisette provider instead of prompting for
  provisioning data.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-18 19:49:24 -04:00
  2afd5307a4
• sdk: fix cross-import overlays with SwiftBuild (#361)

  ## What does this PR do?
  
  Fixes #360 cross-import overlays on Linux for builds with swift 6.4+. 
  
  This PR adds "-Xfrontend", "-enable-cross-import-overlays" to
  `toolchan-swb.json` as it was added to `toolchan.json`.
  Also bumps sdkEpoch so SDK will rebuild after update.
  
  ## How was it tested?
  
  Build was tested locally with branch built from sources.
  
  `xtool new example`, then replace the app file
  
  Fix was tested on this example:
  ```Swift
  import SwiftUI
  import PhotosUI
  
  @main
  struct exampleApp: App {
      @State private var item: PhotosPickerItem?
  
      var body: some Scene {
          WindowGroup {
              PhotosPicker("Pick a photo", selection: $item)
          }
      }
  }
  ```
  
  It fails on version 1.22.2, but builds normally when xtool built from
  this branch.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Configuration Updates**
  * Generated Swift toolchain settings no longer explicitly select the LLD
  linker. Existing frontend and cross-import overlay options remain
  unchanged. This affects linker selection for generated toolsets; no
  public API changes are included.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Sergey Grechishnikov · 2026-10-18 01:36:45 -04:00
  c3c958d760
• certs: match certificates whose serial starts with 0 (#306)

  ## What does this PR do?
  
  Fixes #305 (thanks @0x53 for the detailed diagnosis).
  
  `Certificate.serialNumber()` stripped leading `0` *characters* from the
  hex string, so a serial like `0337…` became `337…` and never matched the
  `serialNumber` returned by the API. xtool then treated its stored
  certificate as missing: free teams are prompted to revoke a valid
  certificate on every run, and paid teams take the same path and try to
  create another certificate, which @McoreD confirmed fails with `409 …
  You already have a current Development certificate or a pending
  certificate request.`
  
  The character-level `drop` predates b6a40ca, which switched to
  fixed-width `%02hhX` formatting; since then it removes the significant
  zero of any first byte below `0x10`. swift-certificates already strips
  leading zero bytes (including DER sign padding) when parsing, so the
  `drop` isn't needed for padding.
  
  - `serialNumber()` now drops only leading zero *bytes*, matching the API
  and `openssl x509 -serial`.
  - The stored-certificate lookup uses a new
  `Certificate.hasSerialNumber(_:)`, which compares serials numerically
  (ignoring case and leading zeros), so it doesn't depend on Apple's exact
  formatting.
  - Profile creation no longer finds the signing certificate with
  `filter[serialNumber]`. The API lists serials with the leading zero
  (`0E41…`), but the filter only matches them without it (`E41…`), so once
  `serialNumber()` kept the zero the filter matched nothing and profile
  creation failed with a 409
  ([details](https://github.com/xtool-org/xtool/pull/306#issuecomment-6181157860)).
  It now fetches all pages of certificates and matches them with
  `hasSerialNumber(_:)`. Fix by @McoreD, with pagination added so a
  certificate on a later page isn't missed.
  
  ## How was it tested?
  
  - Added `Tests/XToolTests/CertificateTests.swift`: builds a cert with
  the serial from #305, round-trips it through DER, and checks
  `serialNumber()` and `hasSerialNumber(_:)`, plus a high-bit serial so
  DER sign padding doesn't leak into the string. The leading-zero test
  fails on `main` (returns `33742570D6F9684A23311711956A084`) and passes
  with this change.
  - `swift test` (17 tests) and `make lint` pass on macOS (Xcode 27, Swift
  6.4, `XTL_CI=1`).
  - @McoreD tested on Linux (Arch, Swift 6.4.1, Xcode 27 SDK, on top of
  1.22.2) with a paid team whose certificate serial starts with `0`: the
  stored certificate is reused, and provisioning, signing and install to
  an iPhone (iOS 27) succeed end to end (tested before pagination was
  added). `CertificateTests` also pass on Linux.
  - Not tested with a free team.
  
  To generate test certificates, the test target now depends on `Crypto`
  and `X509`, which XKit already uses.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Certificate serial numbers now preserve leading zero nibbles while
  omitting unnecessary padding.
  * Certificate matching now ignores letter case and leading zeros, and
  profile lookups find matching certificates even when the API’s
  serial-number filter treats leading-zero values differently.
  * Certificate lookups continue to use the existing fallback when no
  matching certificate is found.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  ---------
  
  Co-authored-by: McoreD <McoreD@users.noreply.github.com>
  Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>

  Chintan Patel · 2026-10-15 16:44:32 -04:00
  316a11e523
• fix: strip trailing carriage return in console prompt on crlf terminals (#328)

  ### Summary
  In `Console.prompt`, incoming line input from standard input is split on
  `\n` without stripping any preceding carriage return (`\r`). On Windows
  environments or CRLF terminals, user input terminates with `\r\n`,
  leaving a trailing `\r` on decoded input strings.
  
  When downstream helper methods such as `Console.chooseNumber` attempt to
  parse numerical selections using `Int(prompt(...))`, the trailing `\r`
  (e.g., `"1\r"`) causes the conversion to return `nil`, trapping the
  terminal in an infinite prompt loop.
  
  ### Changes
  - In `Console.prompt` (`Sources/XToolSupport/Console.swift`), stripped
  any trailing `\r` from the decoded string before returning.
  - Verified that all interactive helper methods (`promptRequired`,
  `getPassword`, `chooseNumber`, `choose`, and `confirm`) delegate their
  standard input reading to `Console.prompt`, ensuring all input paths are
  protected against CRLF leakages.
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Console input no longer includes a trailing carriage return; all other
  input remains unchanged.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Bit Wolf · 2026-10-14 00:44:37 -04:00
  b50173f7b1
• http: add newer Apple Root CAs (#340)

  ## What does this PR do?
  
  Adds Apple G2 & G3 Root CAs to Linux trust root. Fixes #339.
  
  Apple now uses gsd.apple.com for part of the auth flow, which uses the
  G3 root.
  
  ## How was it tested?
  
  Tried to log in on Linux. Failed before, works after.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Improved TLS certificate trust on Linux by recognizing Apple Root CA,
  G2, and G3 certificates when verifying secure connections. Existing TLS
  validation behavior remains unchanged.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-13 20:11:18 -04:00
  65bf3cc5a1
• gsa: fix login using mismatched client-info (#337)

  ## What does this PR do?
  
  Fixes #329 (error `-80045` on gsa login).
  
  We were sending client info generated from Android Anisette, but
  claiming to be a Mac. Seems like GrandSlam no longer likes this
  mismatch.
  
  ## How was it tested?
  
  - [x] Run `xtool ds certificates` after updating -> logged out,
  provisioning reset
  - [x] Re-auth works
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Updates**
  * Authentication now identifies the client as a Pixel 7 running Android
  14 rather than using a macOS-style client description.
  * The provider ID is now consistent across platforms. Existing
  provisioning may be reset if its saved provider ID differs.
  * Before password login, a notice explains the emulated device and
  environment, recommends considering a secondary Apple ID, and says
  credentials are sent only to Apple.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-13 18:06:11 -04:00
  fae1981275
• packer: custom SwiftVersion parser (#303)

  ## What does this PR do?
  
  Drops `mxcl/Version` in favour of a lighter parser that's specifically
  designed to parse the Swift version.
  
  ## How was it tested?
  
  - [x] Unit tests
  - [x] CI
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **New Features**
  * Swift versions can now be compared by major, minor, and patch number,
  and displayed in a consistent three-part format.
  * Version parsing now accepts two- or three-number versions, including
  supported compiler-output formats with optional prefixes or a `-dev`
  suffix.
  * Versions with an omitted patch number are treated as equivalent to the
  same version with patch `0`.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-06 18:54:38 -04:00
  d802c719e4
• packer: use XcodeProjectFormat for generate-xcode-project (#300)

  ## What does this PR do?
  
  Improves `xtool dev generate-xcode-project`. Instead of constructing an
  XcodeGen project, we directly construct an xcodeproj (with the new
  xcproj format) using
  [apple/xcode-project-format](https://github.com/apple/xcode-project-format).
  For now, we still rely on
  [Tuist/XcodeProj](https://github.com/tuist/XcodeProj) to translate it
  back into a pbxproj for maximal compatibility, but this is treated as an
  implementation detail.
  
  Also adds support for resources and icons in the generated Xcode project
  
  ## How was it tested?
  
  - [x] Generated a simple project, opened it, and ran the app
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  * **New Features**
  * Added support for creating and saving Xcode projects and workspaces in
  `.pbxproj` and `.xcproj` formats.
  * Added tools for opening, converting, and writing project and workspace
  files, including workspace contents.
  * Project generation can target a specified directory, and the command
  reports where the project was created.
    * Project generation is available on non-macOS platforms.
  * Generated projects include product settings and references for
  packages, resources, icons, and app extensions.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-06 18:19:30 -04:00
  8f0cc0adf1
• planner: skip macro targets (#208)

  Currently, when using a package with macros like TCA the build fails
  with the following message.
  ```
  Error: Could not find package containing product 'SwiftSyntaxMacros'
  ```
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Package planning now skips macro targets when collecting resources and
  resolving dependencies. This prevents dependencies belonging to macro
  targets from being traversed as part of planning, while leaving
  dependency handling for other target types unchanged.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  Co-authored-by: Kabir Oberai <oberai.kabir@gmail.com>

  Alex Pawlowski · 2026-10-06 03:14:43 +00:00
  491fe2fc2a
• deps: update swiftlint (#299)

  ## What does this PR do?
  
  Updates SwiftLint to v0.65.1
  
  ## How was it tested?
  
  CI
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  * **Chores**
  * Removed inactive legacy networking implementations; active app
  behavior and user-facing capabilities are unchanged.
  * Updated the configured SwiftLint version and improved selection of its
  Linux download for supported processor architectures. macOS continues to
  use its existing download.
  * No changes to app features or user workflows are included in this
  update.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-10-06 02:15:13 +00:00
  6f87f8f39b
• tests: derive CFBundleSupportedPlatforms from the build triple (#289)

  ## What does this PR do?
  
  Application packing now sets CFBundleSupportedPlatforms from
  Packer.cfBundleSupportedPlatforms(for:). The helper splits the triple on
  hyphens, takes the OS token after apple with any numeric version
  stripped, and returns iPhoneSimulator for an ios triple that also
  contains simulator, iPhoneOS for other ios triples, MacOSX for macos or
  macosx, and iPhoneOS when the triple is missing or unrecognized.
  PackerTests covers those triples. Package.resolved drops the
  swift-issue-reporting pin and adds xctest-dynamic-overlay 1.13.1.
  
  Application builds aimed at the iOS simulator still ship an Info.plist
  whose CFBundleSupportedPlatforms value is iPhoneOS. Hosts that read that
  key treat the archive as a physical-device build and refuse to launch it
  on a simulator. While packing an application, Packer assigns
  CFBundleSupportedPlatforms the fixed array iPhoneOS. The build triple is
  already available on buildSettings, but that assignment never reads it,
  so a simulator triple is packed the same way as a device triple.
  
  Fixes #191
  
  ## How was it tested?
  
  The package builds successfully.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  AI was used for assistance.
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Application metadata now reports the supported platform based on the
  build target, including iOS devices, iOS simulators, and macOS. Unknown
  targets retain the iOS device platform as a fallback.
  * **Tests**
  * Added coverage for platform metadata across supported and unknown
  build targets.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->
  
  Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>

  Matt Van Horn · 2026-10-04 19:33:04 +00:00
  8041b9286b
• package: bump deps (#281)

  ## What does this PR do?
  
  Bumps pointfree deps and explicitly disables OpenCombine so that we
  don't have different dependency graphs when building xtool on Darwin vs
  Linux hosts.
  
  ## How was it tested?
  
  A bunch of `swift package resolve`s
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  * **Chores**
  * Updated supporting package versions and refreshed command-line and
  file-operation handling.
  * These changes do not add or remove app features or alter its public
  interface. No action is needed to continue using the app.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-09-25 02:11:36 -04:00
  720618701c
• sdk: prune Xcode when installing from xip (#280)

  ## What does this PR do?
  
  We only need `Xcode.app/Contents/{Info.plist,version.plist,Developer}`
  for the SDK (and expect that to continue being the case). We already
  delete other directories when using `xtool sdk install Xcode.app` but we
  weren't doing this on `xtool sdk install Xcode.xip`. Prune on the xip
  path as well.
  
  Also prune on update (and bump the SDK epoch) so that existing users
  benefit from this.
  
  ## How was it tested?
  
  - [x] Install SDK from xip, confirm `Xcode.app/Contents` only contains
  those three things
  - [x] Install SDK from app, confirm `Xcode.app/Contents` only contains
  those three things + we don't modify the original app
  - [x] Build SDK from app (`xtool sdk build Xcode.app`), confirm we don't
  clobber the input Xcode app
  - [x] Update SDK, confirm that extra contents are removed
  - [x] SDK auto-updates after updating xtool binary
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * SDK builds from Xcode app and XIP inputs now check for required Xcode
  components before installation. Missing components produce an error.
  * Extra direct items in Xcode’s `Contents` folder are removed during XIP
  builds and app updates. Slim builds remain unchanged.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-09-23 01:14:56 -04:00
  f9706d900b
• sdk: get clang resources from swift toolchain (#274)

  ## What does this PR do?
  
  When building the SDK, derives the clang resource dir from the installed
  Swift toolchain (`swiftc -print-target-info`) instead of `clang
  -print-resource-dir`.
  
  Speculatively fixes #273.
  
  ## How was it tested?
  
  Validated that I can still build on Linux
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Improved detection of the host Clang resource directory by using Swift
  compiler target information, helping ensure more reliable compiler setup
  across supported environments.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-09-21 20:01:02 -04:00
  df1b6fa5e9
• sdk: update darwin-tools and OAM (#270)

  ## What does this PR do?
  
  Bumps darwin-tools and OpenAppleMacros
  
  - Updates LLD to support arm64e.x1 to fix #269
  - Adds FoundationModelsMacros to fix #213
  
  ## How was it tested?
  
  - [x] Confirm that we can build with an Xcode 27 release SDK on Linux
  - [x] Confirm that FoundationModels APIs work
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  
  ## Summary by CodeRabbit
  
  * **Bug Fixes**
  * Improved Darwin SDK tool discovery during Swift builds, including
  support for locating `dsymutil`.
  
  * **Updates**
    * Updated bundled Darwin tool dependencies.
  * Added Foundation Models macros support to the OpenAppleMacros library
  set.
  
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-09-21 01:37:14 -04:00
  916cdffba8
• anisette: use XADI on macOS (#259)

  ## What does this PR do?
  
  Updates macOS build to use the fully-local XADI as the Anisette provider
  instead of Omnisette. Includes a one-time migration, logging the user
  out on macOS when they upgrade.
  
  ## How was it tested?
  
  General:
  
  - [x] Log in and confirm that there's no calls to omnisette
  - [x] `xtool dev` works
  
  Migration:
  
  - [x] Log into Apple ID using an old version of xtool
  - [x] Migrate to new xtool, run `xtool dev build -s` and confirm it
  fails due to being logged out
  - [x] Log in on new xtool, confirm that we stay logged in
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  * **New Features**
  * Authentication now supports tracking the anisette provider used to
  create Xcode tokens.
  * Xcode authentication tokens are automatically cleared when the
  provider changes, helping ensure users sign in again with current
  credentials.
  * Anisette provider identification is supported across macOS and Linux.
  
  * **Bug Fixes**
  * Improved authentication error messages, including clearer messaging
  for logout-related errors.
  * Prevented outdated Xcode authentication tokens from being reused after
  a provider change.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-09-20 03:04:52 -04:00
  eacc35da98
• sdk: SwiftBuild support (#229)

  ## What does this PR do?
  
  Adds support for SwiftBuild on Swift 6.4.
  
  We now point SwiftBuild to iPhoneOS.platform and rely on the real
  [SWBApplePlatform](https://github.com/swiftlang/swift-build/tree/b30ee3c00abf34857084c26bb873c3da06348134/Sources/SWBApplePlatform).
  We also do a little futzing to convince SWB to use our Darwin-compatible
  toolset (lld, librarian, resource dirs). We explicitly _don't_ pass in a
  Swift SDK because that would downgrade SWB to using one of the generic
  platform plugins instead of ApplePlatform.
  
  This also gets us a lot closer to parity with Xcode's behavior when
  building for Apple platforms; we get the same build flags and
  everything. We can probably achieve identical bundling too, if we take
  control of the PIF.
  
  One other change: since we need to pass custom env vars and such, we
  can't rely on sourcekit-lsp's built-in SwiftPM integration anymore. We
  thus create a custom BSP server that trampolines to SwiftPM's build
  server with the right flags set. This effectively makes SwiftPM an
  implementation detail, so we can start messing with the PIF.
  
  ## How was it tested?
  
  - [x] Build with Swift 6.3 on macOS
  - [x] Build with Swift 6.4 on macOS
  - [x] Build with Swift 6.3 on Linux
  - [x] Build with Swift 6.4 on Linux
  - [x] BSP/LSP file creation works
  - [x] LSP support works in VSCode
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  - **New Features**
  - Added a `dev build-server` command for Build Server Protocol
  integration.
  - Projects automatically select the appropriate Swift build system when
  supported.
  - Packaging supports output layouts for Swift Package Manager and Swift
  Build.
    - Added options to skip language-server configuration when needed.
  
  - **Documentation**
  - Updated project templates and tutorials to use `.bsp/xtool.json` for
  IDE features such as syntax highlighting and documentation lookup.
  
  - **Improvements**
  - Enhanced SDK generation and compatibility across supported toolchains
  and platforms.
  - Improved automatic language-server configuration while preserving
  existing configuration files.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-09-20 02:47:12 -04:00
  0108899e04
• sdk: integrate OpenAppleMacros (#101)

  ## What does this PR do?
  
  Adds support for proprietary Apple macros (like SwiftUI's `@Entry` and
  `@State`) via
  [OpenAppleMacros](https://github.com/xtool-org/OpenAppleMacros).
  
  We have to re-implement these because the macros need to run on the
  _host_, and the Apple SDKs include them as macOS-only dylibs.
  
  ## How was it tested?
  
  - [x] Extensive integration tests in OpenAppleMacros that compare
  Apple's expansions against ours
  - [x] Confirmed that `@Entry`, `#Preview`, and `@State` work in Swift
  6.3 and 6.4 (State is a PW in the former, macro in the latter)
  - [x] Confirmed that OSS toolchain macros continue to work (tested
  `@Observable`, `#bundle`)
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->

  Kabir Oberai · 2026-09-19 04:28:51 -04:00
  317e018f2b
• packer: keep swift build stdout on stdout (#261)

  per #260 discussion.

  Anton · 2026-09-19 03:28:58 -04:00
  7b97cdd4dd
• sdk: self-updating darwin tools (#110)

  ## What does this PR do?
  
  Allows xtool to auto-update the Darwin SDK format when there's a new
  Darwin toolset, or when we change the SDK format. Downstream this also
  allows us to auto-update when there's changes to OpenAppleMacros (#101).
  
  One callout: the disk footprint of `darwin.artifactbundle` is now ~7 GB
  instead of ~3 GB.
  
  Reason: `xtool sdk install /path/to/Xcode.[app|xip]` now performs a
  "normal" install by default, which preserves the entirety of
  `Xcode.app/Contents/Developer` — even the bits we don't currently use,
  like `AppleTVOS.platform`. This makes it so if we start using more
  components in the future, the user doesn't need to download Xcode again
  and rebuild the SDK themselves.
  
  You can still build an old-style SDK (containing just the components we
  need) using `xtool sdk install --slim` or the dedicated `xtool sdk
  build`.
  
  ## How was it tested?
  
  Tested all of the install modes. Confirmed that auto-updating works.
  Added integration tests in #252.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->
  
  
  <!-- This is an auto-generated comment: release notes by coderabbit.ai
  -->
  ## Summary by CodeRabbit
  
  - **New Features**
  - Added SDK flavor reporting for slim, normal, and legacy installations.
  - Added `sdk update` support to rebuild and install SDKs from an
  installed Xcode bundle.
    - Added a `--slim` option for SDK installation.
    - SDK status now displays installation path, flavor, and version.
  - SDK setup and build operations now ensure a suitable SDK is available
  automatically.
  
  - **Bug Fixes**
  - Improved temporary SDK bundle handling, validation, replacement, and
  cross-filesystem copying.
    - Copy failures now provide detailed diagnostic information.
  <!-- end of auto-generated comment: release notes by coderabbit.ai -->

  Kabir Oberai · 2026-09-18 23:26:10 -04:00
  6efd77fc0e
• anisette: use XADI Swift Package (#258)

  ## What does this PR do?
  
  The xadi repo (https://github.com/xtool-org/xadi) now vends an
  artifactbundle for xadi, so we don't need to separately build/install it
  during the xtool build. This also makes it super easy to support macOS
  later (#259).
  
  ## How was it tested?
  
  - [x] Confirmed I can sign in (including provisioning)
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->

  Kabir Oberai · 2026-09-14 19:45:41 -04:00
  c4d8c42f9a
• sdk: move DarwinSDK into PackLib (#256)

  ## What does this PR do?
  
  Mechanical move of DarwinSDK from XToolSupport -> PackLib. Needed for
  https://github.com/xtool-org/xtool/pull/229.
  
  ## How was it tested?
  
  Compiled + CI
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->

  Kabir Oberai · 2026-09-10 18:37:51 -04:00
  4208c77c81
• gsa: fix auth again (#257)

  ## What does this PR do?
  
  Fixes HTTP 503 on GsService… again (followup from #244)
  
  Seems like GSA has stopped accepting `com.apple.dt.Xcode` in the client
  info string. Use `akd` instead.
  
  ## How was it tested?
  
  - [x] Logging in before failed
  - [x] Logging in after succeeds
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [x] **0** - No AI was used to write code
  - [ ] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->

  Kabir Oberai · 2026-09-10 18:37:31 -04:00
  450223eb07
• sdk: fix permission issue when copying clang resources (#255)

  ## What does this PR do?
  
  Fixes #254 by using `cp -R` instead of `FileManager.copyItem`. The
  latter tries to preserve the original owner of files instead of using
  the caller, which fails if you're non-root and the original files are
  root-owned.
  
  ## How was it tested?
  
  I was able to repro the original issue by installing an SDK as a
  non-root user in an xtool Docker container.
  
  Applied this patch, tried `xtool sdk install` again, and it worked.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->

  Kabir Oberai · 2026-09-10 17:23:15 -04:00
  d4f6443f55
• http: add some debugging knobs (#249)

  ## What does this PR do?
  
  Improves support for HTTP MITM
  
  ## How was it tested?
  
  Did some MITMing on macOS + Linux via Docker, using Proxyman and
  mitmproxy. Didn't test `XTL_HTTP_PROXY` but Sol pinkie promises me that
  it's correct.
  
  ## AI tool usage
  
  How much of this PR was AI-assisted? (check one)
  
  - [ ] **0** - No AI was used to write code
  - [x] **1** - I was assisted by AI. I reviewed the finished result.
  - [ ] **2** - I set the AI going and left it to it; nobody has read the
  result - no review, or AI review only
  
  <!-- If an AI agent is filling this in: declare the level honestly, and
  open as a draft if it is #2. Do not lower the declared level to get the
  PR reviewed. -->

  Kabir Oberai · 2026-09-09 06:50:47 +00:00
  b49941a62a
• ci: enable warnings as errors (#248)
  Kabir Oberai · 2026-09-09 01:49:00 -04:00
  189654c235
• gsa: disregard keepalive (#247)

  GSA seems to, sometimes, reject HTTP requests made on a kept-alive
  connection with a 503. This was either the real root cause, or a second
  root cause, for #243. Fix this by creating a new HTTP connection for
  each GSA request.
  
  See also: https://github.com/rileytestut/AltSign/pull/52 and
  https://github.com/rileytestut/AltSign/pull/47
  
  Speculatively fixes #232

  Kabir Oberai · 2026-09-08 21:23:04 -04:00
  8c77ca7f0b
• http: throw on unsuccessful responses (#246)

  Ref #243. `makeRequest` never inspected `response.status`, so a non-2xx
  body was
  returned as if it had succeeded. Every caller decodes that body as a
  property list
  or JSON, so Apple answering with an HTML error page surfaced as:
  
  ```
  Error: DecodingError.dataCorrupted: Data was corrupted. Debug description: The given
  data was not a valid property list.. Underlying error: ... Encountered unknown tag
  html on line 1
  ```
  
  which says nothing about what actually happened. In the case that
  prompted #243, GSA
  was returning `503 Service Temporarily Unavailable` for the `apptokens`
  operation. It
  now reads:
  
  ```
  Error: POST https://gsa.apple.com/grandslam/GsService2 failed: HTTP 503 Service
  Unavailable (text/html): <html> <head><title>503 Service Temporarily
  Unavailable</title></head> ...
  ```
  
  ### Notes
  
  - The response body is only included in the message for `text/html` and
  `text/plain`,
  i.e. gateway error pages. Bodies from the auth endpoints can carry
  account details
  (the 2FA requests use `Accept: application/x-buddyml`), and errors get
  printed
  verbatim in places such as `DevCommand.swift:164`, so those are left
  out. It also
    avoids emitting replacement characters for binary bodies.
  - The non-2xx path does not report progress, so `XADIProvider`'s
  download no longer
    prints `100%` immediately before failing.
  - Body collection moved to `collect(_:onProgress:)` so there is a single
  exit point.
    Buffer pre-sizing and progress reporting are unchanged.
  - `SDKBuilder` already does this check by hand on `send()`, so it is
  unaffected.
  
  ### Behaviour changes worth a look
  
  - `XADIProvider` writes the Apple Music APK download straight to disk; a
  failed
    download now throws instead of leaving a corrupt `applemusic.apk`.
  - The 2FA requests (`GrandSlamTrustedDeviceRequest`,
  `GrandSlamSMSAuthRequest`,
  `GrandSlamValidateRequest`) use no-op decoders, so a rejected request is
  currently
  discarded and xtool goes on to prompt for a code that was never sent —
  the symptom
    in #235. Those now fail loudly.
  
  ### Two things I left alone
  
  - `GrandSlamTwoFactorAuthenticateOperation.swift:47` catches
  `GrandSlamOperationError`
  with code `-21669` to report an incorrect verification code, but that
  error is only
  thrown by `GrandSlamOperationDecoder`, which the validate requests do
  not use, so the
  catch is unreachable. With this change a wrong code surfaces as a raw
  HTTP error
  instead. Mapping the status back to `incorrectVerificationCode` seemed
  like your call,
    since I do not know what GSA returns for a bad code.
  - If any endpoint reports errors in a decodable body under a non-2xx
  status, this guard
  would pre-empt it. I could not find one — `developerservices2` returns
  HTTP 200 with
    `resultCode`/`userString` in the body, which is untouched here.
  
  Built and linted on Linux (`swift build --product xtool`, `make lint`: 0
  violations).
  
  ---------
  
  Co-authored-by: Robetronica <noreply@Robetronica.com>
  Co-authored-by: Kabir Oberai <oberai.kabir@gmail.com>

  Rob · 2026-09-08 16:48:11 -04:00
  76b005f6d8