I regularly help developers with keychain problems, both here on DevForums and for my Day Job™ in DTS. Many of these problems are caused by a fundamental misunderstanding of how the keychain works. This post is my attempt to explain that. I wrote it primarily so that Future Quinn™ can direct folks here rather than explain everything from scratch (-: If you have questions or comments about any of this, put them in a new thread and apply the Security tag so that I see it. Share and Enjoy — Quinn “The Eskimo!” @ Developer Technical Support @ Apple let myEmail = "eskimo" + "1" + "@" + "apple.com" SecItem: Fundamentals or How I Learned to Stop Worrying and Love the SecItem API The SecItem API seems very simple. After all, it only has four function calls, how hard can it be? In reality, things are not that easy. Various factors contribute to making this API much trickier than it might seem at first glance. This post explains the fundamental underpinnings of the keychain. For information about specific issues, see its companion post, SecItem: Pitfalls and Best Practices. Keychain Documentation Your basic starting point should be Keychain Items. If your code runs on the Mac, also read TN3137 On Mac keychain APIs and implementations. Read the doc comments in <Security/SecItem.h>. In many cases those doc comments contain critical tidbits. When you read keychain documentation [1] and doc comments, keep in mind that statements specific to iOS typically apply to iPadOS, tvOS, and watchOS as well (r. 102786959). Also, they typically apply to macOS when you target the data protection keychain. Conversely, statements specific to macOS may not apply when you target the data protection keychain. [1] Except TN3137, which is very clear about this (-: Caveat Mac Developer macOS supports two different keychain implementations: the original file-based keychain and the iOS-style data protection keychain. IMPORTANT If you’re able to use the data protection keychain, do so. It’ll make your life easier. See the Careful With that Shim, Mac Developer section of SecItem: Pitfalls and Best Practices for more about this. TN3137 On Mac keychain APIs and implementations explains this distinction. It also says: The file-based keychain is on the road to deprecation. This is talking about the implementation, not any specific API. The SecItem API can’t be deprecated because it works with both the data protection keychain and the file-based keychain. However, Apple has deprecated many APIs that are specific to the file-based keychain, for example, SecKeychainCreate. TN3137 also notes that some programs, like launchd daemons, can’t use the file-based keychain. If you’re working on such a program then you don’t have to worry about the deprecation of these file-based keychain APIs. You’re already stuck with the file-based keychain implementation, so using a deprecated file-based keychain API doesn’t make things worse. The Four Freedoms^H^H^H^H^H^H^H^H Functions The SecItem API contains just four functions: SecItemAdd(_:_:) SecItemCopyMatching(_:_:) SecItemUpdate(_:_:) SecItemDelete(_:) These directly map to standard SQL database operations: SecItemAdd(_:_:) maps to INSERT. SecItemCopyMatching(_:_:) maps to SELECT. SecItemUpdate(_:_:) maps to UPDATE. SecItemDelete(_:) maps to DELETE. You can think of each keychain item class (generic password, certificate, and so on) as a separate SQL table within the database. The rows of that table are the individual keychain items for that class and the columns are the attributes of those items. Note Except for the digital identity class, kSecClassIdentity, where the values are split across the certificate and key tables. See Digital Identities Aren’t Real in SecItem: Pitfalls and Best Practices. This is not an accident. The data protection keychain is actually implemented as an SQLite database. If you’re curious about its structure, examine it on the Mac by pointing your favourite SQLite inspection tool — for example, the sqlite3 command-line tool — at the keychain database in ~/Library/Keychains/UUU/keychain-2.db, where UUU is a UUID. WARNING Do not depend on the location and structure of this file. These have changed in the past and are likely to change again in the future. If you embed knowledge of them into a shipping product, it’s likely that your product will have binary compatibility problems at some point in the future. The only reason I’m mentioning them here is because I find it helpful to poke around in the file to get a better understanding of how the API works. For information about which attributes are supported by each keychain item class — that is, what columns are in each table — see the Note box at the top of Item Attribute Keys and Values. Alternatively, look at the Attribute Key Constants doc comment in <Security/SecItem.h>. Uniqueness A critical part of the keychain model is uniqueness. How does the keychain determine if item A is the same as item B? It turns out that this is class dependent. For each keychain item class there is a set of attributes that form the uniqueness constraint for items of that class. That is, if you try to add item A where all of its attributes are the same as item B, the add fails with errSecDuplicateItem. For more information, see the errSecDuplicateItem page. It has lists of attributes that make up this uniqueness constraint, one for each class. These uniqueness constraints are a major source of confusion, as discussed in the Queries and the Uniqueness Constraints section of SecItem: Pitfalls and Best Practices. Parameter Blocks Understanding The SecItem API is a classic ‘parameter block’ API. All of its inputs are dictionaries, and you have to know which properties to set in each dictionary to achieve your desired result. Likewise for when you read properties in output dictionaries. There are five different property groups: The item class property, kSecClass, determines the class of item you’re operating on: kSecClassGenericPassword, kSecClassCertificate, and so on. The item attribute properties, like kSecAttrAccessGroup, map directly to keychain item attributes. The search properties, like kSecMatchLimit, control how the system runs a query. The return type properties, like kSecReturnAttributes, determine what values the query returns. The value type properties, like kSecValueRef perform multiple duties, as explained below. There are other properties that perform a variety of specific functions. For example, kSecUseDataProtectionKeychain tells macOS to use the data protection keychain instead of the file-based keychain. These properties are hard to describe in general; for the details, see the documentation for each such property. Inputs Each of the four SecItem functions take dictionary input parameters of the same type, CFDictionary, but these dictionaries are not the same. Different dictionaries support different property groups: The first parameter of SecItemAdd(_:_:) is an add dictionary. It supports all property groups except the search properties. The first parameter of SecItemCopyMatching(_:_:) is a query and return dictionary. It supports all property groups. The first parameter of SecItemUpdate(_:_:) is a pure query dictionary. It supports all property groups except the return type properties. Likewise for the only parameter of SecItemDelete(_:). The second parameter of SecItemUpdate(_:_:) is an update dictionary. It supports the item attribute and value type property groups. Outputs Two of the SecItem functions, SecItemAdd(_:_:) and SecItemCopyMatching(_:_:), return values. These output parameters are of type CFTypeRef because the type of value you get back depends on the return type properties you supply in the input dictionary: If you supply a single return type property, except kSecReturnAttributes, you get back a value appropriate for that return type. If you supply multiple return type properties or kSecReturnAttributes, you get back a dictionary. This supports the item attribute and value type property groups. To get a non-attribute value from this dictionary, use the value type property that corresponds to its return type property. For example, if you set kSecReturnPersistentRef in the input dictionary, use kSecValuePersistentRef to get the persistent reference from the output dictionary. In the single item case, the type of value you get back depends on the return type property and the keychain item class: For kSecReturnData you get back the keychain item’s data. This makes most sense for password items, where the data holds the password. It also works for certificate items, where you get back the DER-encoded certificate. Using this for key items is kinda sketchy. If you want to export a key, called SecKeyCopyExternalRepresentation. Using this for digital identity items is nonsensical. For kSecReturnRef you get back an object reference. This only works for keychain item classes that have an object representation, namely certificates, keys, and digital identities. You get back a SecCertificate, a SecKey, or a SecIdentity, respectively. For kSecReturnPersistentRef you get back a data value that holds the persistent reference. Value Type Subtleties There are three properties in the value type property group: kSecValueData kSecValueRef kSecValuePersistentRef Their semantics vary based on the dictionary type. For kSecValueData: In an add dictionary, this is the value of the item to add. For example, when adding a generic password item (kSecClassGenericPassword), the value of this key is a Data value containing the password. This is not supported in a query dictionary. In an update dictionary, this is the new value for the item. For kSecValueRef: In add and query dictionaries, the system infers the class property and attribute properties from the supplied object. For example, if you supply a certificate object (SecCertificate, created using SecCertificateCreateWithData), the system will infer a kSecClass value of kSecClassCertificate and various attribute values, like kSecAttrSerialNumber, from that certificate object. This is not supported in an update dictionary. For kSecValuePersistentRef: For query dictionaries, this uniquely identifies the item to operate on. This is not supported in add and update dictionaries. Revision History 2025-05-28 Expanded the Caveat Mac Developer section to cover some subtleties associated with the deprecation of the file-based keychain. 2023-09-12 Fixed various bugs in the revision history. Added a paragraph explaining how to determine which attributes are supported by each keychain item class. 2023-02-22 Made minor editorial changes. 2023-01-28 First posted.

Hello, I’m working on a security agent plugin for Mac. The plugin provides a mechanism with custom UI via SFAuthorizationPluginView and a privileged mechanism with the business logic. The plugin needs to support unlocking the device, so I changed the authorize right to invoke my agent: <?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>class</key> <string>evaluate-mechanisms</string> <key>created</key> <real>731355374.33196402</real> <key>mechanisms</key> <array> <string>FooBar:loginUI</string> <string>builtin:reset-password,privileged</string> <string>FooBar:authenticate,privileged</string> <string>builtin:authenticate,privileged</string> </array> <key>modified</key> <real>795624943.31730103</real> <key>shared</key> <true/> <key>tries</key> <integer>10000</integer> <key>version</key> <integer>1</integer> </dict> </plist> I also changed the system.login.screensaver right to use authorize-session-owner: <?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>class</key> <string>rule</string> <key>comment</key> <string>The owner or any administrator can unlock the screensaver, set rule to "authenticate-session-owner-or-admin" to enable SecurityAgent.</string> <key>created</key> <real>731355374.33196402</real> <key>modified</key> <real>795624943.32567298</real> <key>rule</key> <array> <string>authenticate-session-owner</string> </array> <key>version</key> <integer>1</integer> </dict> </plist> I also set screenUnlockMode to 2, as was suggested in this thread: macOS Sonoma Lock Screen with SFAutorizationPluginView is not hiding the macOS desktop. In the Display Authorization plugin at screensaver unlock thread, Quinn said that authorization plugins are not able to use Touch ID. However, on a MacBook with at touch bar, when I lock the screen, close the lid, and then open it, the touch bar invites me to unlock with Touch ID. If I choose to do so, the screen unlocks and I can interact with the computer, but the plugin UI stays on screen and never goes away, and after about 30 seconds the screen locks back. I can reliably reproduce it on a MacBook Pro with M1 chip running Tahoe 26.1. Is this a known macOS bug? What can I do about it? Ideally, I would like to be able to integrate Touch ID into my plugin, but since that seems to be impossible, the next best thing would be to reliably turn it off completely. Thanks in advance.

I’m developing an iOS application and aiming to install a PKCS#12 (.p12) certificate into the com.apple.token keychain access group so that Microsoft Edge for iOS, managed via MDM/Intune, can read and use it for client certificate authentication. I’m attempting to save to the com.apple.token keychain access group, but I’m getting error -34018 (errSecMissingEntitlement) and the item isn’t saved. This occurs on both a physical device and the simulator. I’m using SecItemAdd from the Security framework to store it. Is this the correct approach? https://developer.apple.com/documentation/security/secitemadd(::) I have added com.apple.token to Keychain Sharing. I have also added com.apple.token to the app’s entitlements. Here is the code I’m using to observe this behavior: public static func installToTokenGroup(p12Data: Data, password: String) throws -> SecIdentity { // First, import the P12 to get the identity let options: [String: Any] = [ kSecImportExportPassphrase as String: password ] var items: CFArray? let importStatus = SecPKCS12Import(p12Data as CFData, options as CFDictionary, &items) guard importStatus == errSecSuccess, let array = items as? [[String: Any]], let dict = array.first else { throw NSError(domain: NSOSStatusErrorDomain, code: Int(importStatus), userInfo: [NSLocalizedDescriptionKey: "Failed to import P12: \(importStatus)"]) } let identity = dict[kSecImportItemIdentity as String] as! SecIdentity let addQuery: [String: Any] = [ kSecClass as String: kSecClassIdentity, kSecValueRef as String: identity, kSecAttrLabel as String: kSecAttrAccessGroupToken, kSecAttrAccessible as String: kSecAttrAccessibleAfterFirstUnlock, kSecAttrAccessGroup as String: kSecAttrAccessGroupToken ] let status = SecItemAdd(addQuery as CFDictionary, nil) if status != errSecSuccess && status != errSecDuplicateItem { throw NSError(domain: NSOSStatusErrorDomain, code: Int(status), userInfo: [NSLocalizedDescriptionKey: "Failed to add to token group: \(status)"]) } return identity }

I'm working on securing communication between an Authorization Plugin and an XPC daemon, and I’d appreciate some guidance on best practices and troubleshooting. The current design which, I’ve implemented a custom Authorization Plugin for step-up authentication, which is loaded by Authorization Services at the loginwindow (inside SecurityAgent). This plugin acts as an XPC client and connects to a custom XPC daemon. Setup Details 1. XPC Daemon Runs as root (LaunchDaemon) Not sandboxed (my understanding is that root daemons typically don’t run sandboxed—please correct me if this is wrong) Mach service: com.roboInc.AuthXpcDaemon Bundle identifier: com.roboInc.OfflineAuthXpcDaemon 2. Authorization Plugin Bundle identifier: com.roboInc.AuthPlugin Loaded by SecurityAgent during login 3. Code Signing Both plugin and daemon are signed using a development certificate What I’m Trying to Achieve I want to secure the XPC communication so that: The daemon only accepts connections from trusted clients The plugin only connects to the legitimate daemon Communication is protected against unauthorized access The Issue I'm facing I attempted to validate code signatures using: SecRequirementCreateWithString SecCodeCopyGuestWithAttributes SecCodeCheckValidity However, validation consistently fails with: -67050 (errSecCSReqFailed) Could you please help here What is the recommended way to securely authenticate an Authorization Plugin (running inside SecurityAgent) to a privileged XPC daemon? Since the plugin runs inside SecurityAgent, how can the daemon reliably distinguish my plugin from other plugins? What is the correct approach to building a SecRequirement in this scenario? Any guidance, examples, or pointers would be greatly appreciated. Thanks in advance!

We have successfully deployed our Qt C++ application on Windows and Android using OpenSSL with TLS Pre-Shared Key (PSK) authentication to connect to our servers. However, I understand that apps submitted to the App Store must use SecureTransport as the TLS backend on iOS. My understandiunig is that SecureTransport does not support PSK ciphersuites, which is critical for our security architecture. Questions: Does SecureTransport support TLS PSK authentication, or are there plans to add this feature? If PSK is not supported, what is Apple's recommended alternative for applications that require PSK-based authentication? Is there an approved exception process that would allow me to use OpenSSL for TLS connections on iOS while still complying with App Store guidelines? The application requires PSK for secure communication with our infrastructure, and we need guidance on how to maintain feature parity across all platforms while meeting App Store requirements

I downloaded a P12 file (containing a private key) from the company server, and retrieved the private key from this P12 file using a password : private func loadPrivateKeyFromPKCS12(path: String, password: String) throws -> SecKey? { let p12Data: Data do { p12Data = try Data(contentsOf: fileURL) } catch let readError { ... } let options: [CFString: Any] = [ kSecImportExportPassphrase: password as CFString ] var items: CFArray? let status = SecPKCS12Import(p12Data as CFData, options as CFDictionary, &items) guard status == errSecSuccess else { throw exception } var privateKey: SecKey? let idd = identity as! SecIdentity let _ = SecIdentityCopyPrivateKey(idd, &privateKey) return privateKey } However, when I use this private key to call SecKeyCreateSignature for data signing, a dialog box always pops up to ask user to input the Mac admin password. What confuses me is that this private key is clearly stored in the local P12 file, and there should be no access to the keychain involved in this process. Why does the system still require the user's login password for signing? Is it possible to perform silent signing (without the system dialog popping up) in this scenario?

I am developing a macOS system service (standalone binary running as a LaunchDaemon) that requires the ability to sign data using a private key which will be deployed via MDM. The Setup: Deployment: A .mobileconfig pushes a PKCS12 identity to the System Keychain. Security Requirement: For compliance and security reasons, we cannot set AllowAllAppsAccess to <true/>. The key must remain restricted. The Goal: I need to use the private key from the identity to be able to sign the data The Problem: The Certificate Payload does not support a TrustedApplications or AccessControl array to pre-authorize binary paths. As a result, when the process tries to use the private key for signing (SecKeyCreateSignature), it prompts the user to allow this operation which creates a disruption and is not desired. What i've tried so far: Manually adding my process to the key's ACL in keychain access obviously works and prevents any prompts but this is not an "automatable" solution. Using security tool in a script to attempt to modify the ACL in an automated way, but that also asks user for password and is not seamless. The Question: Is there a documented, MDM-compatible way to inject a specific binary path into the ACL of a private key? If not, is there a better way to achieve the end goal?

Hello there, Starting from iOS 18.4, support was included for QWAC Validation and QCStatements. Using the official QWAC Validator at: https://eidas.ec.europa.eu/efda/qwac-validation-tool I was able to check that the domain "eidas.ec.europa.eu" has a valid QWAC certificate. However, when trying to obtain the same result using the new API, I do not obtain the same result. Here is my sample playground code: import Foundation import Security import PlaygroundSupport PlaygroundPage.current.needsIndefiniteExecution = true @MainActor class CertificateFetcher: NSObject, URLSessionDelegate { private let url: URL init(url: URL) { self.url = url super.init() } func start() { let session = URLSession(configuration: .ephemeral, delegate: self, delegateQueue: nil) let task = session.dataTask(with: url) { data, response, error in if let error = error { print("Error during request: \(error)") } else { print("Request completed.") } } task.resume() } nonisolated func urlSession(_ session: URLSession, didReceive challenge: URLAuthenticationChallenge, completionHandler: @escaping (URLSession.AuthChallengeDisposition, URLCredential?) -&gt; Void) { guard let trust = challenge.protectionSpace.serverTrust else { completionHandler(.cancelAuthenticationChallenge, nil) return } if let certificates = SecTrustCopyCertificateChain(trust) as? [SecCertificate] { self.checkQWAC(certificates: certificates) } let credential = URLCredential(trust: trust) completionHandler(.useCredential, credential) } nonisolated func checkQWAC(certificates: [SecCertificate]) { let policy = SecPolicyCreateSSL(true, nil) var trust: SecTrust? guard SecTrustCreateWithCertificates(certificates as CFArray, policy, &amp;trust) == noErr, let trust else { print("Unable to create SecTrust") return } var error: CFError? guard SecTrustEvaluateWithError(trust, &amp;error) else { print("Trust evaluation failed") return } guard let result = SecTrustCopyResult(trust) as? [String : Any] else { print("No result dictionary") return } let qwacStatus = result[kSecTrustQWACValidation as String] let qcStatements = result[kSecTrustQCStatements as String] print("QWAC Status: \(String(describing: qwacStatus))") print("QC Statements: \(String(describing: qcStatements))") } } let url = URL(string: "https://eidas.ec.europa.eu/")! let fetcher = CertificateFetcher(url: url) fetcher.start() Which prints: QWAC Status: nil QC Statements: nil Request completed. Am I making a mistake while using the Security framework? I would greatly appreciate any help or guidance you can provide.

We have a custom SecurityAgentPlugin that is triggered by multiple authorizationdb entries. Some customers report that the SecurityAgent process takes window focus even though no UI or windows are displayed. Our plugin explicitly ignores the _securityAgent user and does not show any UI for that user. However, in macOS 26.1, it appears that the plugin still causes the SecurityAgent to take focus as soon as it is triggered. Is this a change in macOS 26.1 or a bug? Can we do anything to prevent "focus stealing"?

Since the introduction of the siblings / and /System/Volumes/Data architecture, some very basic, critical commands seems to have a broken behaviour ( cp, rsync, tar, cpio…). As an example, ditto which was introduced more than 10 years ago to integrate correctly all the peculiarity of HFS Apple filesystem as compared to the UFS Unix filesystem is not behaving correctly. For example, from man ditto: --rsrc Preserve resource forks and HFS meta-data. ditto will store this data in Carbon-compatible ._ AppleDouble files on filesystems that do not natively support resource forks. As of Mac OS X 10.4, --rsrc is default behavior. [...] --extattr Preserve extended attributes (requires --rsrc). As of Mac OS X 10.5, --extattr is the default. and nonetheless: # ls -@delO /private/var/db/ConfigurationProfiles/Store drwx------@ 5 root wheel datavault 160 Jan 20 2024 /private/var/db/ConfigurationProfiles/Store                            ********* com.apple.rootless 28 *************************** # mkdir tmp # ditto /private/var/db/ConfigurationProfiles tmp ditto: /Users/alice/Security/Admin/Apple/APFS/tmp/Settings: Operation not permitted ditto: /Users/alice/Security/Admin/Apple/APFS/tmp/Store: Operation not permitted # ls -@delO tmp/Store drwx------ 5 root wheel - 160 Aug 8 13:55 tmp/Store                            * # The extended attribute on copied directory Store is empty, the file flags are missing, not preserved as documented and as usual behaviour of ditto was since a long time ( macOS 10.5 ). cp, rsync, tar, cpio exhibit the same misbehaviour. But I was using ditto to be sure to avoid any incompatibility with the Apple FS propriaitary modifications. As a consequence, all backup scripts and applications are failing more or less silently, and provide corrupted copies of files or directories. ( I was here investigating why one of my security backup shell script was making corrupted backups, and only on macOS ). How to recover the standard behaviour --extattr working on modern macOS?

Hello, We are facing an issue with performing a DTLS handshake when our iOS application is in the background. Our app (Vocera Collaboration Suite – VCS) uses secure DTLS-encrypted communication for incoming VoIP calls. Problem Summary: When the app is in the background and a VoIP PushKit notification arrives, we attempt to establish a DTLS handshake over our existing socket. However, the handshake consistently fails unless the app is already in the foreground. Once the app is foregrounded, the same DTLS handshake logic succeeds immediately. Key Questions: Is performing a DTLS handshake while the app is in the background technically supported by iOS? Or is this an OS-level limitation by design? If not supported, what is the Apple-recommended alternative to establish secure DTLS communication for VoIP flows without bringing the app to the foreground? Any guidance or clarification from Apple engineers or anyone who has solved a similar problem would be greatly appreciated. Thank you.

I am not very well versed in this area, so I would appreciate some guidance on what should be enabled or disabled. My app is an AppKit app. I have read the documentation and watched the video, but I find it hard to understand. When I added the Enhanced Security capability in Xcode, the following options were enabled automatically: Memory Safety Enable Enhanced Security Typed Allocator Runtime Protections Enable Additional Runtime Platform Restrictions Authenticate Pointers Enable Read-only Platform Memory The following options were disabled by default: Memory Safety Enable Hardware Memory Tagging Memory Tag Pure Data Prevent Receiving Tagged Memory Enable Soft Mode for Memory Tagging Should I enable these options? Is there anything I should consider disabling?

I am curious as to know if i calculate the checksum of an ipa file and upload the same to app store, and then after installing the app on my device, if i extract the ipa file and compare the checksum will it match? or will it vary from device to device, because of bitcode and app thinning slicing? Some banks have been showing ipa file checksums on their websites, and even inside their apps and showing messages like checksum matches! i was just curious as to know how would one go about validating this!? Or is this even possible, what about the checksum of the executable at runtime? Can we check this? will it match?

iOS mTLS Client Certificate Authentication Fails in TestFlight with Error -25303 Problem I'm building an iOS app that uses mTLS (client certificates received from server at runtime). Storing SecCertificate to keychain fails with error -25303 in both development and TestFlight builds, preventing SecIdentity creation needed for URLSession authentication. Environment: iOS 18.2, iPad Pro, TestFlight internal testing, keychain-access-groups properly configured Diagnostic Results Testing keychain operations shows an interesting pattern: ✅ Generic Password - Works: let addQuery: [CFString: Any] = [ kSecClass: kSecClassGenericPassword, kSecAttrAccount: "test", kSecValueData: "password".data(using: .utf8)! ] SecItemAdd(addQuery as CFDictionary, nil) // Returns: 0 (success) ✅ SecKey - Works: let addKeyQuery: [CFString: Any] = [ kSecClass: kSecClassKey, kSecValueRef: privateKey, kSecAttrApplicationTag: tag ] SecItemAdd(addKeyQuery as CFDictionary, nil) // Returns: 0 (success) ❌ SecCertificate - Fails: let addCertQuery: [CFString: Any] = [ kSecClass: kSecClassCertificate, kSecValueRef: certificate, // Created from server-provided PEM kSecAttrApplicationTag: tag ] SecItemAdd(addCertQuery as CFDictionary, nil) // Returns: -25303 Code Context Attempting to create SecIdentity for mTLS: private func createIdentity(fromCert certPEM: String, key keyPEM: String) throws -> SecIdentity { // 1. Parse PEM to DER and create SecCertificate - succeeds guard let certData = extractPEMData(from: certPEM, type: "CERTIFICATE"), let certificate = SecCertificateCreateWithData(nil, certData as CFData) else { throw CertificateError.invalidCertificate } // 2. Parse PEM key and create SecKey - succeeds guard let keyData = extractPEMData(from: keyPEM, type: "PRIVATE KEY"), let privateKey = SecKeyCreateWithData(keyData as CFData, attrs as CFDictionary, &error) else { throw CertificateError.invalidKey } // 3. Add key to keychain - SUCCEEDS (errSecSuccess) let tempTag = UUID().uuidString.data(using: .utf8)! SecItemAdd([ kSecClass: kSecClassKey, kSecValueRef: privateKey, kSecAttrApplicationTag: tempTag ] as CFDictionary, nil) // ✅ Works // 4. Add certificate to keychain - FAILS (-25303) let status = SecItemAdd([ kSecClass: kSecClassCertificate, kSecValueRef: certificate, kSecAttrApplicationTag: tempTag ] as CFDictionary, nil) // ❌ Fails with -25303 guard status == errSecSuccess else { throw CertificateError.keychainError(status) } // 5. Would query for SecIdentity (never reached) // ... } Network Behavior When mTLS fails, console shows: Connection: asked for TLS Client Certificates Connection: received response for client certificates (-1 elements) Connection: providing TLS Client Identity (-1 elements) Task received response, status 403 The -1 elements indicates no certificates were provided. Entitlements <key>keychain-access-groups</key> <array> <string>$(AppIdentifierPrefix)com.ellin.tshios</string> </array> Keychain Sharing capability is enabled. What I've Tried Both kSecValueRef and kSecValueData approaches - same error Various kSecAttrAccessible values - same error Different keychain access groups - same error TestFlight build (vs dev build) - same error PKCS#12 creation - requires complex ASN.1/DER encoding, no iOS API Questions Is error -25303 expected when adding SecCertificate in development/TestFlight builds? Will App Store distribution resolve this? Or is there a fundamental limitation? Why does SecKey succeed but SecCertificate fails with identical entitlements? Is there an alternative to create SecIdentity without keychain access? Constraints Certificates come from server at runtime (cannot bundle) Need SecIdentity for URLSession client certificate authentication Server provides PEM format certificates Tested on: Simulator (dev), iPad Pro (dev), iPad Pro (TestFlight) - all fail Any insights appreciated - specifically whether this is a provisioning profile limitation that App Store distribution would resolve.

Hi, After enabling the new Enhanced Security capability in Xcode 26, I’m seeing install failures on devices running < iOS 26. Deployment target: iOS 15.0 Capability: Enhanced Security (added via Signing & Capabilities tab) Building to iOS 18 device error - Unable to Install ...Please ensure sure that your app is signed by a valid provisioning profile. It works fine on iOS 26 devices. I’d like to confirm Apple’s intent here: Is this capability formally supported only on iOS 26 and later, and therefore incompatible with earlier OS versions? Or should older systems ignore the entitlement, meaning this behavior might be a bug?

(Xcode 26.2, iPhone 17 Pro) I can't seem to get hardware tag checks to work in an app launched without the special "Hardware Memory Tagging" diagnostics. In other words, I have been unable to reproduce the crash example at 6:40 in Apple's video "Secure your app with Memory Integrity Enforcement". When I write a heap overflow or a UAF, it is picked up perfectly provided I enable the "Hardware Memory Tagging" feature under Scheme Diagnostics. If I instead add the Enhanced Security capability with the memory-tagging related entitlements: I'm seeing distinct memory tags being assigned in pointers returned by malloc (without the capability, this is not the case) Tag mismatches are not being caught or enforced, regardless of soft mode The behaviour is the same whether I launch from Xcode without "Hardware Memory Tagging", or if I launch the app by tapping it on launchpad. In case it was related to debug builds, I also tried creating an ad hoc IPA and it didn't make any difference. I realise there's a wrinkle here that the debugger sets MallocTagAll=1, so possibly it will pick up a wider range of issues. However I would have expected that a straight UAF would be caught. For example, this test code demonstrates that tagging is active but it doesn't crash: #define PTR_TAG(p) ((unsigned)(((uintptr_t)(p) >> 56) & 0xF)) void *p1 = malloc(32); void *p2 = malloc(32); void *p3 = malloc(32); os_log(OS_LOG_DEFAULT, "p1 = %p (tag: %u)\n", p1, PTR_TAG(p1)); os_log(OS_LOG_DEFAULT, "p2 = %p (tag: %u)\n", p2, PTR_TAG(p2)); os_log(OS_LOG_DEFAULT, "p3 = %p (tag: %u)\n", p3, PTR_TAG(p3)); free(p2); void *p2_realloc = malloc(32); os_log(OS_LOG_DEFAULT, "p2 after free+malloc = %p (tag: %u)\n", p2_realloc, PTR_TAG(p2_realloc)); // Is p2_realloc the same address as p2 but different tag? os_log(OS_LOG_DEFAULT, "Same address? %s\n", ((uintptr_t)p2 & 0x00FFFFFFFFFFFFFF) == ((uintptr_t)p2_realloc & 0x00FFFFFFFFFFFFFF) ? "YES" : "NO"); // Now try to use the OLD pointer p2 os_log(OS_LOG_DEFAULT, "Attempting use-after-free via old pointer p2...\n"); volatile char c = *(volatile char *)p2; // Should this crash? os_log(OS_LOG_DEFAULT, "Read succeeded! Value: %d\n", c); Example output: p1 = 0xf00000b71019660 (tag: 15) p2 = 0x200000b711958c0 (tag: 2) p3 = 0x300000b711958e0 (tag: 3) p2 after free+malloc = 0x700000b71019680 (tag: 7) Same address? NO Attempting use-after-free via old pointer p2... Read succeeded! Value: -55 For reference, these are my entitlements. [Dict] [Key] application-identifier [Value] [String] … [Key] com.apple.developer.team-identifier [Value] [String] … [Key] com.apple.security.hardened-process [Value] [Bool] true [Key] com.apple.security.hardened-process.checked-allocations [Value] [Bool] true [Key] com.apple.security.hardened-process.checked-allocations.enable-pure-data [Value] [Bool] true [Key] com.apple.security.hardened-process.dyld-ro [Value] [Bool] true [Key] com.apple.security.hardened-process.enhanced-security-version [Value] [Int] 1 [Key] com.apple.security.hardened-process.hardened-heap [Value] [Bool] true [Key] com.apple.security.hardened-process.platform-restrictions [Value] [Int] 2 [Key] get-task-allow [Value] [Bool] true What do I need to do to make Memory Integrity Enforcement do something outside the debugger?

Our Goal: We are implementing a workflow for derived credentials. Our objective is to have a PIV/CAC derived credential (from Entrust), installed via the Intune MDM Company Portal app, and then use it within our (managed) app to generate digital signatures. Challenge: The Intune Company Portal installs these identities into the System Keychain. Because third-party apps are restricted from accessing private keys in the System Keychain, we are running into a roadblock. Our Question: 1) Is there an API that allows us to create a signature without us having to pass the private key itself, but instead just pass a handle/some reference to the private key and then the API can access the private key in the system keychain and create the signature under the hood. SecKeyCreateSignature is the API method that creates a signature but requires passing a private key. 2) If #1 is not feasible, is there a way to get access to system keychain to retrieve certs + private key for managed apps

Hi, I’m seeing a production issue on iOS 26+ that only affects some users. symptoms: It does NOT happen for all users. It happens for a subset of users on iOS 26+. If we write a value to Keychain and read it immediately in the same session, it succeeds. However, after terminating the app and relaunching, the value appears to be gone: SecItemCopyMatching returns errSecItemNotFound (-25300). Repro (as observed on affected devices): Launch app (iOS 26+). Save PIN data to Keychain using SecItemAdd (GenericPassword). Immediately read it using SecItemCopyMatching -> success. Terminate the app (swipe up / kill). Relaunch the app and read again using the same service -> returns -25300. Expected: The Keychain item should persist across app relaunch and remain readable (while the device is unlocked). Actual: After app relaunch, SecItemCopyMatching returns errSecItemNotFound (-25300) as if the item does not exist. Implementation details (ObjC): We store a “PIN” item like this (simplified): addItem: kSecClass: kSecClassGenericPassword kSecAttrService: <FIXED_STRING> kSecValueData: kSecAttrAccessControl: SecAccessControlCreateWithFlags(..., kSecAttrAccessibleWhenUnlockedThisDeviceOnly, 0, ...) readItem (SecItemCopyMatching): kSecClass: kSecClassGenericPassword kSecAttrService: <FIXED_STRING> kSecReturnData: YES (uses kSecUseOperationPrompt in our async method) Question: On iOS 26+, is there any known issue or new behavior where a successfully added GenericPassword item could later return errSecItemNotFound after app termination/relaunch for only some users/devices? What should we check to distinguish: OS behavior change/bug vs. entitlement/access-group differences (app vs extension, provisioning/team changes), device state/policies (MDM, passcode/biometrics changes), query attributes we should include to make the item stable across relaunch? Build / Dev Environment: macOS: 15.6.1 (24G90) Xcode: 26.2

I'm trying to authenticate to a git host using SSH keys stored in 1Password. I have ~/.ssh/config with mode 600 set with a symlink: Host * IdentityAgent "~/.1password/agent.sock" But ssh-add -l shows no identities. If I set $SSH_AUTH_SOCK, ssh-add -l works just fine. I'd love to not have to do this, though. Why doesn't ssh-add seem to read ~/.ssh/config? The built-in version is OpenSSH_10.0p2, LibreSSL 3.3.6. I've searched fruitlessly for an answer anywhere else.

Apple has repeatedly ignored my reports about a critical privacy issue in Spotlight on macOS 26, and the problem persists in version 26.3 RC. This is not a minor glitch, it is a fundamental breach of user trust and privacy. Several aspects of Spotlight fail to respect user settings: • Hidden apps still exposed: In the Apps section (Cmd+1), Spotlight continues to display apps marked with the hidden flag, even though they should remain invisible. • Clipboard reactivation: The clipboard feature repeatedly turns itself back on after logout or restart, despite being explicitly disabled by the user. • Excluded files revealed: Most concerning, Spotlight exposes files in Suggestions and Recents (Cmd+3) even when those files are explicitly excluded under System Settings > Spotlight > Search Privacy. This behavior directly violates user expectations and system settings. It is not only a major privacy issue but also a security risk, since sensitive files can be surfaced without consent. Apple must address this immediately. Users rely on Spotlight to respect their privacy configurations, and the current behavior undermines both trust and security.