# FakeWallet iOS Crypto Stealer Campaign Delivered Through 26 Apple App Store Apps (SparkKitty-linked, Chinese-speaking Actor)

> Kaspersky Securelist disclosed 26 phishing iOS applications distributed through the official Apple App Store — most concentrated in the Chinese storefront — impersonating MetaMask, Ledger, Trust Wallet, Coinbase, TokenPocket, imToken, and Bitpie. The FakeWallet family (Kaspersky: HEUR:Trojan-PSW.IphoneOS.FakeWallet.*) exfiltrates BIP-39 seed phrases by combining dylib library injection, Objective-C method swizzling of viewDidLoad on recovery-phrase screens, view-controller hierarchy traversal, custom __hook Mach-O sections, and React Native payloads. Stolen mnemonics are RSA-PKCS#1 encrypted, Base64-encoded, and POSTed to C2 endpoints on kkkhhhnnn[.]com, helllo2025[.]com, sxsfcc[.]com, iosfc[.]com, nmu8n[.]com, zmx6f[.]com, and api.dc1637[.]xyz. Tooling and campaign infrastructure overlap with the SparkKitty trojan, and log strings plus Chinese-language phishing pages attribute the operation to a Chinese-speaking actor with operator metadata dating to at least fall 2025. Both hot-wallet (silent recovery-phrase scraping) and cold-wallet (in-app phishing WebView prompting the victim to 'verify' their seed phrase) vectors are in active use, placing every crypto user who sideloads one of the impersonator apps at immediate risk of full wallet drain.

- **Published:** 2026-04-20T12:00:00Z
- **Last reviewed:** 2026-04-20T12:00:00Z
- **Canonical:** https://intel.threadlinqs.com/threat/TL-2026-0395
- **Card image:** https://intel.threadlinqs.com/og/threat/threat-intel-fakewallet-crypto-stealer-tl-2026-0395.webp?v=0a16bde2 (Threadlinqs threat intel: FakeWallet iOS Stealer in 26 Apps: Detection & IOCs (TL-2026-0395) — Critical severity, 56 IOCs, 9 detection rules.)
- **ID:** TL-2026-0395
- **Severity:** CRITICAL
- **Category:** MALWARE
- **Status:** MONITORING
- **Actor:** FakeWallet operators (China)
- **Detections:** 9 (1 shown free below) · **IOCs:** 56 (3 shown free below)

## Description

On 20 April 2026, Kaspersky Securelist (Sergey Puzan, GReAT) published a technical deep-dive revealing that throughout March 2026 the Apple App Store — particularly the Chinese storefront — was flooded with phishing applications masquerading as well-known cryptocurrency wallets. Kaspersky identified 26 such apps spanning MetaMask, Ledger, Trust Wallet, Coinbase, TokenPocket, imToken, and Bitpie. Because Apple's regional policies block the legitimate wallet apps for users whose Apple ID is set to the China region, the FakeWallet operators weaponised typosquatting (mimicked icons, intentionally misspelled names) and deceptive promotional banners that claimed the 'official wallet is unavailable in the App Store' to drive installs. Several of the apps have since been pulled following Kaspersky's disclosure to Apple, but additional dormant apps without yet-activated phishing logic remain in the ecosystem and are positioned for a switch-flip via future update.

Each App Store app itself is only a stub — a calculator, task planner, or small game — whose real purpose is to open a browser link on launch. That link pivots the victim through a malicious enterprise provisioning-profile install of a trojanised wallet build. Enterprise provisioning profiles were designed for corporate in-house app distribution, but have long been abused by cracked-app, casino, cheat, and malware operators — and are the same profile type that SparkKitty leveraged. The installed payload then behaves in one of three technical patterns depending on the target wallet.

Pattern 1 — Hot wallet injection via dylib + Objective-C swizzling: The attackers add extra load commands to the main Mach-O executable so the dyld linker pulls in a malicious library (e.g. libokexHook.dylib in the Coinbase clone). Initialisation routines — either Objective-C +[UIViewController load] and +[UIView load] methods, or C++ static initialisers in __mod_init_functions — swap the original viewDidLoad of the recovery-phrase screen (e.g. RecoveryPhraseViewController) with a hijacked version. The hooked viewDidLoad walks the current view controller's subview tree, extracts mnemonic words as they are entered, concatenates them, encrypts the resulting string with RSA using PKCS #1 padding, Base64-encodes the ciphertext, and POSTs it together with a module type, app name, and a hardcoded unique code to a C2 URL of the form /api/open/postByTokenPocket?ciyu=<b64>&code=10001&ciyuType=1&wallet=<name>.

Pattern 2 — Trust Wallet __hook trampoline: Instead of initialisation hooks, the attackers append a custom __hook section to the main executable, placed just before __text in the region normally reserved for load commands. The first two entries are trampolines resolving dlsym and the original WalletCore mnemonic-validation routine; the next pair are wrappers that resolve the symbols dataInit or processX0Parameter from the malicious library, hand control to them, and then execute the original replaced method. This rewires every code path where the app validates, imports, or generates a seed phrase, leaking the mnemonic to the C2 before the legitimate logic runs.

Pattern 3 — Ledger cold-wallet phishing via malicious library or React Native tamper: Because Ledger Live is a UI front for a hardware wallet and cannot silently read recovery phrases, the attackers use in-app phishing. In the library variant, an entry function loads a config file containing C2 URL (e.g. hxxps://iosfc[.]com/ledger/ios/Rsakeycatch.php), a numeric 'code' such as 10001, and a login-url plus login-code. +[UIViewController load] and +[UIView load] overrides traverse the view hierarchy; when a screen named add-account-cta or one containing a $ sign is rendered, the malware identifies the locale from the 'add account' button text, constructs a localised 'security check' phishing notification, and stores it in GlobalVariables. When the victim subsequently opens an account-add or buy/sell screen, the notification surfaces and opens a WebView pointing at a local verify.html bundled inside the app. verify.html mimics the Ledger UI, enforces BIP-39 dictionary validation on the entered words (rejecting anything off-list to raise credibility), offers autocomplete for mnemonic words, and forwards the completed phrase to an Objective-C handler that encrypts it with RSA PKCS #1, Base64-encodes it, and sends it to the C2. The second Ledger variant patches the React Native JavaScript bundle directly — Ledger Live is open-source — inserting two new screens, MnemonicVerifyScreen (embedded in PortfolioNavigator) and PrivateKeyVerifyScreen (embedded in MyLedgerNavigator). The phishing screens only trigger once a hardware wallet is paired, a deliberate anti-analysis guardrail. Three state files are written to the app directory: verify-wallet-status.json (timestamped progress tracker), verify-wallet-config.json (current C2 configuration), and verify-wallet-pending.json (encrypted mnemonics awaiting transmission; cleared via clearPendingMnemonicJob on success). A dedicated exfil thread resumes pending sends if the app is relaunched.

Infrastructure and cross-platform reach: Kaspersky also discovered a phishing website cloning the official Ledger landing page that served the same iOS payloads plus Android APKs — some previously undocumented — linked via enterprise-profile install flows. No Google Play presence was observed; Android distribution is confined to attacker-controlled sites. Several infected iOS apps contained dormant SparkKitty modules alongside the FakeWallet code, a strong link given SparkKitty's own 2025 campaign used the same enterprise-profile/phishing-page-mimicking-App-Store delivery model, shared Chinese-language log strings, and pursued the same cryptocurrency-theft objectives. Kaspersky assesses with medium-to-high confidence that the two campaigns share operators or are closely collaborating.

Victimology: Near-exclusive Chinese-language phishing pages and Chinese App Store prevalence indicate Chinese-speaking crypto users are the primary targets, but the modules contain no locale gating and some variants auto-localise phishing notifications — any iOS user who ends up on one of the phishing redirects or sideloads a trojanised build via enterprise profile is exposed. The financial impact is severe: a single successful exfiltration transfers full BIP-39 entropy and therefore complete control of every account derivable from the seed, across every chain the wallet supports. Defenders protecting high-net-worth users or on-call IR for Web3 / exchange / custody operations should treat any Chinese-region App Store install of a wallet-themed app as presumptively compromised until proven otherwise, and should hunt for the hashes, C2 domains, and behavioural IOCs enumerated below across mobile-management, proxy, and DNS telemetry.

## MITRE ATT&CK

- T1195 Supply Chain Compromise
- T1195.002 Supply Chain Compromise: Compromise Software Supply Chain
- T1189 Drive-by Compromise
- T1566.002 Phishing: Spearphishing Link
- T1204.001 User Execution: Malicious Link
- T1204.002 User Execution: Malicious File
- T1575 Native API
- T1554 Compromise Host Software Binary
- T1546 Event Triggered Execution
- T1574.006 Hijack Execution Flow: Dynamic Linker Hijacking
- T1574 Hijack Execution Flow
- T1027 Obfuscated Files or Information
- T1036.005 Match Legitimate Resource Name or Location
- T1553.006 Subvert Trust Controls: Code Signing Policy Modification
- T1140 Deobfuscate/Decode Files or Information
- T1056.002 Input Capture: GUI Input Capture
- T1555 Credentials from Password Stores
- T1552.001 Unsecured Credentials: Credentials In Files
- T1056 Input Capture
- T1113 Screen Capture
- T1005 Data from Local System
- T1071.001 Application Layer Protocol: Web Protocols
- T1573.002 Encrypted Channel: Asymmetric Cryptography
- T1132.001 Data Encoding: Standard Encoding
- T1568.002 Dynamic Resolution: Domain Generation Algorithms
- T1041 Exfiltration Over C2 Channel
- T1657 Financial Theft

## Free detection (Splunk SPL, Microsoft KQL, Sigma)

**FakeWallet Mnemonic Exfil URL Fingerprint — C2 Phase** (`TL-2026-0395-SPL-003`)

Detects outbound HTTP/HTTPS requests whose query string simultaneously contains the FakeWallet mnemonic-exfil fingerprint: the Mandarin 'ciyu' parameter carrying RSA-PKCS#1+Base64-encrypted BIP-39 seed data, the hardcoded 'code=10001', 'ciyuType', and a 'wallet=' identifier. This pattern is distinctive across every FakeWallet hot-wallet variant (MetaMask/Coinbase/Trust Wallet/TokenPocket/imToken/Bitpie) and is the single most reliable network IOC.

- **Languages:** Splunk SPL, Microsoft KQL, Sigma
- **ATT&CK:** T1041, T1071.001, T1132.001, T1573.002, T1657
- **Data sources:** proxy, zeek_http, web_gateway, firewall
- **Severity:** critical
- **Confidence:** high

### Splunk SPL

```spl
index=proxy sourcetype="proxy:*" (uri_query="*ciyu=*" AND uri_query="*code=10001*" AND uri_query="*ciyuType=*" AND uri_query="*wallet=*")
| eval ioc="FakeWallet-exfil-url-fingerprint"
| stats min(_time) as first_seen max(_time) as last_seen values(uri_path) as paths values(dest) as c2_hosts values(user_agent) as uas count by src src_user ioc
| where count > 0
| table first_seen last_seen src src_user c2_hosts paths uas count ioc
```

### Microsoft KQL

```kql
let c2_domains = dynamic(["kkkhhhnnn.com", "helllo2025.com", "sxsfcc.com", "iosfc.com", "nmu8n.com", "zmx6f.com", "api.dc1637.xyz"]);
DeviceNetworkEvents
| where Timestamp > ago(24h)
| where RemoteUrl has_any (c2_domains) or
        (RemoteUrl has "ciyu=" and RemoteUrl has "code=10001" and RemoteUrl has "ciyuType=" and RemoteUrl has "wallet=")
| project Timestamp, DeviceName, InitiatingProcessAccountName, InitiatingProcessFileName, RemoteIP, RemoteUrl, ActionType
| extend Threat = "FakeWallet-C2-exfil"
```

### Sigma

```yaml
title: FakeWallet Mnemonic Exfil URL Fingerprint
id: 8b1a2e3c-7f40-4d61-9b51-aa0395000003
status: experimental
description: Detects the distinctive FakeWallet C2 exfil URL pattern combining the Mandarin 'ciyu' query parameter, hardcoded code=10001, ciyuType, and wallet identifier — unique to the FakeWallet iOS crypto-stealer family.
references:
    - https://securelist.com/fakewallet-cryptostealer-ios-app-store-2/119482/
author: Threadlinqs Team
date: 2026/04/20
tags:
    - attack.command_and_control
    - attack.exfiltration
    - attack.impact
    - attack.t1041
    - attack.t1071.001
    - attack.t1132.001
    - attack.t1573.002
    - attack.t1657
logsource:
    category: proxy
detection:
    selection:
        cs-uri-query|contains|all:
            - 'ciyu='
            - 'code=10001'
            - 'ciyuType='
            - 'wallet='
    condition: selection
falsepositives:
    - None expected; the four-parameter combination is unique to FakeWallet.
level: critical

```

**Known false positives:**

- None expected — the four-parameter combination is unique to FakeWallet C2. Benign services may use ciyu or wallet individually but never together with code=10001 and ciyuType.

**Tuning notes:** Pivot on src to identify the infected mobile device or tethered workstation; cross-reference with any libokexHook or verify-wallet-*.json file events in the prior 60 minutes. If the proxy truncates query strings, add the C2 domain list (kkkhhhnnn.com, helllo2025.com, sxsfcc.com, iosfc.com, nmu8n.com, zmx6f.com, api.dc1637.xyz) as a secondary selection.

### 8 more detection rules (Splunk SPL, Microsoft KQL, Sigma), unlock with TL Creds

- FakeWallet React Native Bundle Tamper — Exploitation Phase (locked: unlock with TL Creds)
- FakeWallet Multi-Artefact Host Correlation — Post-Exploitation Phase (locked: unlock with TL Creds)
- FakeWallet C2 Domain Contact — Command and Control Phase (locked: unlock with TL Creds)
- FakeWallet Mach-O Static Tamper Signature — Exploitation Phase (locked: unlock with TL Creds)
- FakeWallet Ledger Cold-Wallet Phishing Bundle — Post-Exploitation Phase (locked: unlock with TL Creds)
- FakeWallet RSA+Base64 Mnemonic Exfil POST — C2 Phase (locked: unlock with TL Creds)
- FakeWallet libokexHook Injection Library Drop — Exploitation Phase (locked: unlock with TL Creds)
- FakeWallet Ledger RN State File Creation — Post-Exploitation Phase (locked: unlock with TL Creds)

Sign up free to unlock the detection rules: https://threadlinqs.com/?modal=signup

## Sample indicators (3 of 56)

Defanged, TLP:CLEAR.

- IP address: `139.180.139[.]209`
- Domain: `kkkhhhnnn[.]com`
- URL: `hxxps://api[.]npoint[.]io/153b165a59f8f7d7b097`

Full set of 56 indicators unlocks with TL Creds: https://threadlinqs.com/tl-creds

## Sources

- [FakeWallet crypto stealer spreading through iOS apps in the App Store](https://securelist.com/fakewallet-cryptostealer-ios-app-store-2/119482/)
- [SparkKitty trojan research (Kaspersky)](https://securelist.com/sparkkitty-trojan/)
- [ESET — Trojanized wallets target phones (historical precursor, 2022)](https://www.welivesecurity.com/2022/03/24/crypto-malware-patched-wallets-targeting-android-ios-devices/)
- [Apple Platform Security — Provisioning profile overview](https://support.apple.com/guide/security/welcome/web)
- [BIP-39 Mnemonic code for generating deterministic keys](https://github.com/bitcoin/bips/blob/master/bip-0039.mediawiki)
- [Ledger Live (open-source React Native source, abused as modification base)](https://github.com/LedgerHQ/ledger-live)
- [MITRE ATT&CK for Mobile — T1575 Native API (iOS hooking)](https://attack.mitre.org/techniques/T1575/)

## Full data

The other 8 detection rules (Splunk SPL / Microsoft KQL / Sigma) unlock with TL Creds; a free account includes daily creds: https://threadlinqs.com/?modal=signup
Indicator values unlock with TL Creds, at a lower price on Red and Purple; AI agents can query rules and IOCs through the Threadlinqs MCP server (Purple): https://intel.threadlinqs.com/mcp

Canonical: https://intel.threadlinqs.com/threat/TL-2026-0395
