Hey, Just recently I realized something I have been overlooking in my build pipelines. I thought that by adding the the "hardened runtime", I disable 3rd-party library injection (I do not have the disable-library-validation entitlement added). However, I was using some checks on my code and I noticed that the "library validation" code signature check fails on my applications (e.g. adding the .libraryValidation requirement via the LightweightCodeRequirements framework) - with codesign -dvvvv /path/to/app I can check it doesn't have the CS_REQUIRE_LV flag: [...] CodeDirectory v=20500 size=937 flags=0x10000(runtime) hashes=18+7 location=embedded [...] then I used in Xcode the "Other Code Signing Flags" setting and added the -o library option, which added the flag: [...] CodeDirectory v=20500 size=937 flags=0x12000(library-validation,runtime) hashes=18+7 location=embedded [...] Is this flag something I should be explicitly setting? Because I was under the impression enabling hardened runtime would be enough. Popular Developer ID distributed applications (e.g. Google Chrome, Parallels Desktop, Slack) all have this flag set.

The Developer App Certificate is not trusted.

I have a free developer account, and I have been creating applications. When I tried to open one of them, it said that this app has been flagged as malware. It is not malware, so I don't know why it has been flagged as this. Not just this app, but suddenly a whole bunch of my apps have been flagged as malware as well! The app I have been developing is basically a windows Taskbar for my macbook air, and it has been working well until the latest update i made where it hides in full screen, suddenly it started taking up significant energy, so i reverted to an older version while i was fixing it. Then, when i try to open it another time, it starts to open, and it says "Malware Blocked and Moved to Bin" “Taskbar.app” was not opened because it contains malware. This action did not harm your Mac”. All versions of the taskbar now contain this message. I try opening some of my other apps, a shared storage client and a shared storage server (where i was testing with app groups), and they couldn't open either, the same malware message appeared. ProPermission couldn't open either (changes permissions on files for me so i don't have to use the terminal or finder). I can run these apps through the Xcode environment (attached process), but when I archieve it into an app bundle, the malware flag appears. Please note that I am certain that these apps do not contain malware, apparently XProtect has incorrectly flagged my apps as malware. Because I do not have the paid developer account, I cannot notarize my apps. I am using MacOS Tahoe 26.1 with Xcode 26.0, and I have tested it with a iMac Intel 2017 with MacOS Ventura.

I'm experiencing an issue when exporting an Enterprise distribution certificate where the certificate and private key won't export together - the private key keeps getting left out. I'm running macOS Tahoe. Has anyone encountered the same issue or know of a solution? Any help would be appreciated.

I am using a Sonoma VM (14.6) where i have installed xcode 16.2. When i try login into apple id into xcode, i am getting this error. I know i am entering the correct credentials. Not sure why this issue is. In other Sequioa and Tahoe VM , i was able to login.

Hello Apple Developer Forum Community, I’ve got a problem with the signing process of my AppClip Test App. Can someone help me? As I don’t know hot to get the certificate...

Hello, I sent this in as a feedback several weeks ago about watchOS 26.2 beta 2 but since the issue is still active now that watchOS 26.2 is in production I'm reposting here for the community. I would also like to submit a DTS about this issue but honestly don't know the best way to go about it and would appreciate advice about that. There seems to be an issue with VPP distribution for our app on watchOS 26.2. When our watchOS companion app is launched after being installed through VPP to a supervised iPhone, it encounters a dyld error before main() or any application code is even called. The same app launches correctly in every other circumstance we could imagine and test: – Installed through VPP on supervised devices running watchOS 26.1. – Installed from the app store (using an apple id) on a supervised iPhone and paired watch running iOS 26.2 / watchOS 26.2. – Installed through Testflight on a supervised iPhone and paired watch running iOS 26.2 / watchOS 26.2. – Installed through the app store on unsupervised devices running watchOS 26.1 and 26.2. This strongly appears to be a VPP signing issue because we even did the following experiment: Install iPhone and Watch apps through the App Store on a supervised device pair running public iOS 26.2 beta 2 / watchOS 26.2 beta 2. Verify that both apps launch successfully. Use an MDM command to install from VPP over the existing installations Verify that the watch app fails to launch (the iOS app is unaffected) My feedback included some crash logs which I won't be reposting publicly here. Any feedback or ideas appreciated.

I suddenly started to receive the following email with the error in it stating that my uploaded app is not available to be used in TestFlight: ITMS-90886: 'Cannot be used with TestFlight because the signature for the bundle at “MyApp.app/Contents/PlugIns/MyAppWidgetExtension.appex” is missing an application identifier but has an application identifier in the provisioning profile for the bundle. Bundles with application identifiers in the provisioning profile are expected to have the same identifier signed into the bundle in order to be eligible for TestFlight.' It was all working fine and now I am not sure even where to start looking. Signing, provisioning and everything else is managed automatically.

I am trying to code sign an application which relies on many python libraries to run. For background knowledge, the .app was created with a —onefile command on Visual Studio. I code signed my application itself using codesign --deep --force --verify --timestamp --sign "Developer ID Application: Issey Yohannes (GL5BCCW69X)" /Users/isseyyohannes/Desktop/Automated\ ALGORA.app However, when I try to run the application the error shows in terminal as follows [PYI-16345:ERROR] Failed to load Python shared library '/var/folders/g9/2zbc7y_97xxbq7bnc301nnyc0000gn/T/_MEI6keRcA/Python': dlopen: dlopen(/var/folders/g9/2zbc7y_97xxbq7bnc301nnyc0000gn/T/_MEI6keRcA/Python, 10): no suitable image found. Did find: /var/folders/g9/2zbc7y_97xxbq7bnc301nnyc0000gn/T/_MEI6keRcA/Python: code signature in (/var/folders/g9/2zbc7y_97xxbq7bnc301nnyc0000gn/T/_MEI6keRcA/Python) not valid for use in process using Library Validation: mapped file has no Team ID and is not a platform binary (signed with custom identity or adhoc?) /var/folders/g9/2zbc7y_97xxbq7bnc301nnyc0000gn/T/_MEI6keRcA/Python: stat() failed with errno=1 Through some external tools, I was able to narrow the issue as follows Hardened Runtime Restriction: Your application is attempting to load a shared library (Python) at runtime, but the library is either: Not properly signed with the same Team ID as your app. Not marked as a valid platform binary. macOS requires all loaded libraries to comply with its code-signing and runtime security policies. Any insight is much appreciated.

I've noticed that NSTaks has this property as of macOS 14.4 @property (nullable, copy) NSData *launchRequirementData API_AVAILABLE(macos(14.4)) API_UNAVAILABLE(ios, watchos, tvos, visionos); It has no documentation whatsoever. Even google search has no clue. Does this have anything to do with code signature requirements validation? Any explanations and examples would be appreciated!

Hi everyone. Sorry if this is not an appropriate forum section for this question. I'm making a game engine and it doesn't launch on my colleague's MacBook, although it does launch on mine. There's an application file, let's say, Sample.app. And along with it in the same folder there's Engine.dylib. If we look at the app-file structure, the executable file's path is Contents/MacOS/Sample. So for the executable file the library is located at the path ../../../Engine.dylib. But when my colleague runs the Sample.app file, he gets an error "Library not loaded: @executable_path/../../../Engine.dylib". Although the path is correct and on my MacBook it works. Are there any ideas how to fix it?

Hi, I'm doing Java 22 experiments with open source package "wgpu-macos-aarch64-release" on my Mac M3. This library contains a .dylib file and a .a file. In Java you can load native dynamic libraries using JNI. I used that in the past. Now in MacOS 15 this stopped working. Currently, I compile my Java file and run the binary file on the command line, including the library directory. The application would load the library from that directory. This should normally works but the library is not loaded. After some investigation it seems the .dylib is rejected by MacOS. Calling spctl --assess --verbose=4 libwgpu_native.dylib gives libwgpu_native.dylib: rejected without any extra reason message. Using sudo xattr -rd com.apple.quarantine libwgpu_native.dylib doesn't solve the problem. I also can't "accept" it in "Privacy & Security" because it's run as a command line call. I know that code signing should be done for "complete" applications. But these are just experiments and working with single dynamic libraries. How should this be done, to be able to load those libraries in a Java command line program? Or in general, how can you load an unsigned dylib in an application (apart from using Java as a tool). Kr, J

I am developing a macOS app that requires the Associated Domains entitlement. The app will be distributed as a custom app. The app needs to be signed using Team A’s Developer ID Application certificate and packaged under Team A’s Team ID. Team A has a secure signing and packaging setup, but they do not provide access to their Developer ID Application Identity (cert) or their provisioning profile. I am part of Team B and have access to Team B’s Developer ID Application identity and provisioning profiles. I am thinking of doing the following: I create a provisioning profile under Team B that authorizes the Associated Domains entitlement. I sign the app using Team B’s Developer ID Application identity, ensuring the required entitlements are included. Then, I re-sign the app using Team A’s Developer ID Application identity, since Team A has also set up the same bundle ID with the Associated Domains entitlement and corresponding provisioning profile. Questions: Is this approach correct & does it have any drawback? Will the double signing process work without issues, given that Team A has the required provisioning profile for the same bundle ID? Are there better ways to handle this situation where signing must be done under Team A but access is limited? Thanks!

When I first tried to sign my local unit test with the identity generated by Xcode, it failed because the intermediate certificate was missing. In that case, the error message explained that the trust chain could not be completed. But after installing the correct intermediate, codesign still fails, but no longer gives any explanation: codesign -f -s '0EFE7E591A4E690842094B8EC5AFDFE059637D3C' build/Darwin-Xcode-arm64_obf/bin/Release/UNITTEST build/Darwin-Xcode-arm64_obf/bin/Release/UNITTEST: replacing existing signature build/Darwin-Xcode-arm64_obf/bin/Release/UNITTEST: errSecInternalComponent It's the same error line "errSecInternalComponent". Is there a log somewhere that might explain what exactly is the error?

All libraries are getting rejected with errors like: not valid for use in process: mapping process and mapped file (non-platform) have different Team IDs

The Codesign command adds extended attributes to files that previously had no extended attributes. In my case codesign add following extended attributes to text file in Frrameworks folder: com.apple.cs.CodeDirectory com.apple.cs.CodeRequirements com.apple.cs.CodeRequirements-1 com.apple.cs.CodeSignature Can I somehow prevent this behavior? Thank you.

Hello, I'm working on an app at work and we finally got to signing and notarizing the app. The app is successfully notarized and stapled, I packaged it in a .dmg using hdiutil and went ahead and notarized and stapled that as well. Now I tried to move this app to another machine through various methods. But every time I download it from another machine, open and extract the contents of the dmg and attempt to open the app, I get the "Apple could not verify my app is free of malware that may harm your Mac or compromise your privacy. When I check the extended attributes there's always the com.apple.quarantine attribute which from what I know, is the reason that this popup appears I've tried uploading it to google drive, sending through slack, onedrive, even tried our AWS servers and last but not least, I tried our Azure servers (which is what we use for distribution of the windows version of our app). I tried uploading to Azure through CloudBerry (MSP360 now), and azure-cli defining the content-type as "application/octet-stream", the content-disposition as "attachment; filename=myApp.dmg", and content-cache-control as "no-transform". None of these worked The only times where a download actually worked with no problems was when I downloaded through the terminal using curl, which obviously not a great solution especially that we're distributing to users who aren't exactly "tech savy" I want the installation experience to be as smooth as other apps outside the App Store (i.e Discord, Slack, Firefox, Chrome etc....) but I've been stuck on this for more than a week with no luck. Any help is greatly appreciated, and if you want me to clarify something further I'd be happy to do so

Hello I have a qt, CMAKE app, non-xcode one till xcode start supporting cmake. I have 3 apps, 2 basic ones and 1 very complex ones. My complex one build/links/notarises/validates/deploys beautifly. I have tear in my eye when I see it build. The other 2 apps explode and torment me for past 5 days. The build proves is 99% the same, the only thing that a little changes are info.plist and app name+ some minor changes. Its absolutely bananas and I can't fix it, I'm running out of ideas so if any1 could sugged anything, I'll buy & ship you a beer. Anyway, errors: Log: Using otool: Log: inspecting "/Users/dariusz/Qt/6.9.1/macos/lib/QtCore.framework/Versions/A/QtCore" Log: Could not parse otool output line: "/Users/dariusz/Qt/6.9.1/macos/lib/QtCore.framework/Versions/A/QtCore (architecture arm64):" Log: Adding framework: Log: Framework name "QtCore.framework" Framework directory "/Users/dariusz/Qt/6.9.1/macos/lib/" Framework path "/Users/dariusz/Qt/6.9.1/macos/lib/QtCore.framework" Binary directory "Versions/A" Binary name "QtCore" Binary path "/Versions/A/QtCore" Version "A" Install name "@rpath/QtCore.framework/Versions/A/QtCore" Deployed install name "@rpath/QtCore.framework/Versions/A/QtCore" Source file Path "/Users/dariusz/Qt/6.9.1/macos/lib/QtCore.framework/Versions/A/QtCore" Framework Destination Directory (relative to bundle) "Contents/Frameworks/QtCore.framework" Binary Destination Directory (relative to bundle) "Contents/Frameworks/QtCore.framework/Versions/A" Log: copied: "/Users/dariusz/Qt/6.9.1/macos/lib/QtWebSockets.framework/Versions/A/QtWebSockets" Log: to "/AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/Frameworks/QtWebSockets.framework/Versions/A/QtWebSockets" Log: copy: "/Users/dariusz/Qt/6.9.1/macos/lib/QtWebSockets.framework/Resources" "/AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/Frameworks/QtWebSockets.framework/Versions/A/Resources" Log: copied: "/Users/dariusz/Qt/6.9.1/macos/lib/QtWebSockets.framework/Resources/Info.plist" Log: to "/AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/Frameworks/QtWebSockets.framework/Versions/A/Resources/Info.plist" Log: copied: "/Users/dariusz/Qt/6.9.1/macos/lib/QtWebSockets.framework/Resources/PrivacyInfo.xcprivacy" Log: to "/AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/Frameworks/QtWebSockets.framework/Versions/A/Resources/PrivacyInfo.xcprivacy" Log: symlink "/AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/Frameworks/QtWebSockets.framework/QtWebSockets" Log: points to "Versions/Current/QtWebSockets" Log: symlink "/AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/Frameworks/QtWebSockets.framework/Resources" Log: points to "Versions/Current/Resources" Log: symlink "/AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/Frameworks/QtWebSockets.framework/Versions/Current" Log: points to "A" Log: Using install_name_tool: Log: in "/AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/MacOS/Agameri_Toolbox" Log: change reference "@rpath/QtWebSockets.framework/Versions/A/QtWebSockets" Log: to "@rpath/QtWebSockets.framework/Versions/A/QtWebSockets" ERROR: "/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/install_name_tool: fatal error: file not in an order that can be processed (link edit information does not fill the __LINKEDIT segment): /AppBuild/Agameri_Toolbox/Agameri_Toolbox.app/Contents/MacOS/Agameri_Toolbox\n" ERROR: "" Even tho I get that error, it will "Notarize" and "greenlight by gatekeeper. So my automatic build app if he sees error with the __LINKEDIT it will stop deployment. But even tho he "stops" release of app to public I can still go check binary and when I try to run that I get: Library not loaded: @rpath/QtConcurrent.framework/Versions/A/QtConcurrent Referenced from: <69A296DB-8C7D-3BC9-A8AE-947B8D6ED224> /Volumes/VOLUME/*/Agameri_Toolbox.app/Contents/MacOS/Agameri_Toolbox Reason: tried: '/Users/dariusz/Qt/6.9.1/macos/lib/QtConcurrent.framework/Versions/A/QtConcurrent' (code signature in <192D5FAC-FE8C-31AB-86A7-6C2CE5D3E864> '/Users/dariusz/Qt/6.9.1/macos/lib/QtConcurrent.framework/Versions/A/QtConcurrent' not valid for use in process: mapping process and mapped file (non-platform) have different Team IDs), '/System/Volumes/Preboot/Cryptexes/OS/Users/dariusz/Qt/6.9.1/macos/lib/QtConcurrent.framework/Versions/A/QtConcurrent' (no such file), '/Volumes/DEV_MAC/02_CODE/Dev/Icarus.nosync/Icarus_Singleton/codeSingleton/libOutput/Release/QtConcurrent.framework/Versions/A/QtConcurrent' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/DEV_MAC/02_CODE/Dev/Icarus.nosync/Icarus_Singleton/codeSingleton/libOu (terminated at launch; ignore backtrace) And here is my build script, its QT based application, I'm using macdeployqt + my own custom signing as the one from macdeployqt breaks on the complex app. (I will post it in next post as apparently there is 7k limit O.O) I've tried to replace the @rpath/ to @executable_path but that has made a million new issues and I'm just lost.

The driver does not show up in the app settings after switching to “DriverKit USB Transport - VendorID”. Previously, the app used “DriverKit USB Transport (development)” and everything worked as expected. The entitlements looked like this: <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> <plist version="1.0"> <dict> <key>com.apple.developer.driverkit</key> <true/> <key>com.apple.developer.driverkit.transport.usb</key> <array> <dict> <key>idVendor</key> <string>*</string> </dict> </array> </dict> </plist> I received approval to use “DriverKit USB Transport - VendorID”. I updated the App ID configuration in the portal, removed all development entitlements, updated the provisioning profile, and edited the driver’s .entitlements as follows: <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> <plist version="1.0"> <dict> <key>com.apple.developer.driverkit</key> <true/> <key>com.apple.developer.driverkit.transport.usb</key> <array> <dict> <key>idVendor</key> <integer>1111</integer> </dict> </array> </dict> </plist> The app installs on an iPad with an M processor, but the driver does not appear in the settings. In the logs I see the following: 272 debug 19:50:42.005193+0300 installd 7935 signing bytes in 5 blob(s) from /var/installd/Library/Caches/com.apple.mobile.installd.staging/temp.bugkAE/extracted/Payload/****.app/SystemExtensions/****Driver.dext/****.Driver(arm64) 272 debug 19:50:42.012068+0300 installd open(/var/installd/Library/Caches/com.apple.mobile.installd.staging/temp.bugkAE/extracted/Payload/****.app/SystemExtensions/net.svedm.****.SDRDriver.dext/Info.plist,0x0,0x1b6) = 4 272 debug 19:50:42.012712+0300 installd 0xc2e14c618 done serializing <?xml version="1.0" encoding="UTF-8"?><!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "https://www.apple.com/DTDs/PropertyList-1.0.dtd"><plist version="1.0"><dict><key>application-identifier</key><string>****.Driver</string><key>com.apple.application-identifier</key><string>****</string><key>com.apple.developer.driverkit</key><true/><key>com.apple.developer.driverkit.transport.usb</key><array><dict><key>idVendor</key><integer>3034</integer></dict></array><key>com.apple.developer.team-identifier</key><string>****</string><key>com.apple.security.get-task-allow</key><true/><key>get-task-allow</key><true/></dict></plist> 0 error 19:53:08.930054+0300 kernel Sandbox: MyApp(844) deny(1) sysctl-read kern.bootargs 0 error 19:53:08.931571+0300 kernel Sandbox: driverkitd(77) deny(1) syscall-unix 284 syscall-unix-denied-SIGKILL 0 error 19:53:09.985946+0300 kernel 1 duplicate report for Sandbox: driverkitd(77) deny(1) syscall-unix 284 syscall-unix-denied-SIGKILL 0 error 19:53:09.985985+0300 kernel Sandbox: MyApp(844) deny(2) file-test-existence /usr/bin/swift-backtrace 0 error 19:53:09.986011+0300 kernel Sandbox: MyApp(844) deny(2) file-test-existence /usr/bin/arm64e But I don’t quite understand what is going wrong. Any ideas?

We recently had an external pentest for one of our company's macOS applications and they brought up the topic of library validation. Our app has hardened runtime enabled and passes notarization. The codesign verification output includes: flags=0x10000(runtime) The pentesters brought up that both validation and runtime should be present, so I discovered that you could also add library validation by augmenting our flags with: OTHER_CODE_SIGN_FLAGS = --timestamp -o library which changes the flags to: flags=0x12000(library-validation,runtime) The pentesters insist that both options are necessary, especially to avoid library injection when SIP is off, but Apple's docs say that hardened runtime already implies library validation (see here ) My question is: does explicitly specifying library validation provide something that hardened runtime does not already? Or is it correct that hardened runtime already imply library validation? For what it's worth, I did a quick scan of some of the apps on my system, interesting some of the Apple system apps have only library validation (e.g. Safari, Photos), some have both (e.g. Podcasts), some have only hardened runtime (e.g. Mail). So that didn't help answer the question. Thank you!