Join our Newsletter — 33% off our NHI Course
Home› Guides› Mobile T1453 Abuse Accessibility Features: Detecting Data Collection…
Guide Endpoint Security

Mobile T1453 Abuse Accessibility Features: Detecting Data Collection on Android

PDF for newsletter subscribers · free to read online

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 12 November 2026 10 min read
On this page

Abuse Accessibility Features (T1453) is Android malware persuading the user to switch on an accessibility service, then using the power that feature gives to people with disabilities to read the screen, capture what is typed and tap through the device on the user's behalf. MITRE ATT&CK for Mobile v20 lists T1453 under Collection and Credential Access, for Android only. It is the engine behind most modern Android banking trojans and a growing number of spyware families. This guide is for mobile security, MDM and fraud teams: what the technique is, why ATT&CK does not list iOS, which platform controls and policies apply, ATT&CK's v20 detection analytic and how to respond. It goes deeper than the MITRE ATT&CK for Mobile Guide.

Key takeaways

  • Android only, by ATT&CK's definition. T1453 lists Android as its only platform and every procedure example is Android malware.
  • One user approval unlocks everything. Once a malicious accessibility service is enabled, ATT&CK records families granting themselves more permissions, preventing uninstall, stealing credentials and one-time passcodes.
  • Authenticator codes are in scope. ATT&CK records GodFather exfiltrating Google Authenticator one-time passwords, and ThreatFabric reports Crocodilus reading the Google Authenticator screen on command.
  • Android has added friction. Restricted settings on Android 13 and later lock accessibility for some installed apps until the user explicitly allows it, and Google Play restricts use of the AccessibilityService API. Malware droppers try to get around both.
  • Enterprises can allowlist accessibility services. On managed Android devices, restrict accessibility services to an approved list and alert when an unapproved service is enabled.

What T1453 covers and how it relates to other techniques

ATT&CK describes adversaries abusing accessibility features in Android to steal sensitive data and to spread malware to other devices. It gives two examples from FluBot research: overlaying an HTML object that mimics a login screen so the user types credentials into it, and a malicious accessibility service acting as a keylogger by watching changes to text input fields. ATT&CK notes that keylogging this way is also described in Keylogging (T1417.001), while T1453 captures the overall abuse of accessibility features. GUI Input Capture (T1417.002) and Input Injection (T1516) are separate techniques that accessibility abuse often enables.

What an enabled service gives the malwareEffect recorded by ATT&CK or the cited source
Reading screen content and eventsCrocodilus monitors accessibility events and captures on-screen text; VajraSpy intercepts WhatsApp, WhatsApp Business and Signal communication
Watching which app is in the foregroundCherryBlos monitors when a wallet application launches to steal credentials; Crocodilus shows overlays when target apps launch
Acting on the user's behalfChameleon automatically grants additional permissions and switches biometric authentication to PIN; VajraSpy enables permissions on the user's behalf; Anubis executes screen taps without the victim's knowledge
Defending itselfGodFather and Anubis prevent the user from uninstalling the app; FluBot evades detection and removal
SpreadingFluBot sends SMS for lateral movement

Android and iOS: why the platforms differ

ATT&CK lists T1453 for Android only and records no iOS procedure examples. On Android, an app can provide an accessibility service that receives events from other apps once the user enables it in Settings, under Accessibility. That design is what malware abuses. iOS is not listed for this technique, so this page does not describe an iOS equivalent. iOS collection of screen content and input is covered under other ATT&CK techniques, such as Input Capture (T1417), and in practice relies on elevated privileges rather than a user-granted accessibility service.

ControlAndroid detailWho applies it
User enablementAccessibility services are enabled by the user in Settings, Accessibility; ATT&CK's user guidance points users there to review which apps have itUser, MDM
Restricted settingsGoogle describes restricted settings that lock settings such as accessibility for an installed app until the user chooses "Allow restricted settings" in the app's info page; some steps work only on Android 13 and laterPlatform
Google Play policyOnly apps that help people with disabilities may declare isAccessibilityTool; other apps using the API must show a prominent in-app disclosure, get affirmative consent and complete a declaration in Play ConsoleGoogle Play
Managed device policyOn Android Enterprise, the device policy controller can limit which accessibility services may be enabled (NHI Mgmt Group practice; confirm the setting name in your MDM)Enterprise

Restricted settings and Play policy raise the bar but do not close the route. ThreatFabric reported in March 2025 that Crocodilus is installed through a dropper that bypasses Android 13 and later restrictions, so treat the platform controls as one layer. See the Android Banking Trojans Guide for the wider family landscape.

Real cases recorded by ATT&CK

  • FluBot (S1067): abuses accessibility in three ways: stealing app credentials, evading detection and removal, and sending SMS to spread.
  • Crocodilus (S9004): requests Accessibility Service, then connects to C2 for instructions and monitors accessibility events. ThreatFabric's report, published on 28 March 2025, describes early campaigns against banks in Spain and Turkey and cryptocurrency wallets, an accessibility logger that records on-screen text, and a black overlay used to hide remote control.
  • Chameleon (S1083): switches biometric authentication to PIN, grants itself permissions and prevents uninstall after accessibility is granted.
  • GodFather (S1231): prevents uninstall, exfiltrates Google Authenticator one-time passwords and steals credentials.
  • VajraSpy (S9006) and DocSwap (S9005): spyware use, from intercepting messaging apps to keylogging.

The common thread is a social engineering step that gets the user to enable the service, often with an app disguised as something familiar. ThreatFabric's Crocodilus sample was disguised as a browser.

Detection: ATT&CK v20 analytic

ATT&CK v20 attaches DET0697 Detection of Abuse Accessibility Features to T1453, with a single Android analytic. It correlates three sources:

  • Dynamic analysis: an accessibility service receiving events from more than a threshold number of distinct apps, including apps on a target list such as banking, wallet, messaging and authenticator apps. ATT&CK notes that broad monitoring across apps distinguishes malware from single-purpose assistive tools.
  • Permission state: permission grant state from the device policy controller and the app's requested permissions, reviewed in app vetting.
  • Network traffic: gateway or proxy inspection where it is deployed, to tie the app to C2 or exfiltration.

Operationally, the strongest signals for most enterprises are an MDM or MTD report that an unapproved accessibility service is enabled, an app installed from outside the managed store, and an app that requested accessibility shortly after install. For banks and other app owners, in-app risk signals that an accessibility service is active while a sensitive screen is shown are worth feeding to fraud scoring. See the Mobile Banking App Security Guide.

Prevention

ControlSourceHow to apply it
User guidanceATT&CK M1011Teach users to be wary of suspicious texts and links, of apps asking for accessibility, and to check Settings, Accessibility; ATT&CK also warns that apps on Google Play may not be benign
Managed app sourcesNHI Mgmt Group practiceBlock unknown sources on corporate devices so restricted settings and Play review apply
Accessibility allowlistNHI Mgmt Group practicePermit only approved assistive apps on managed devices and alert on others
Phishing-resistant MFANHI Mgmt Group practiceCodes shown on screen can be read; FIDO2 and passkeys do not expose a reusable code
Recent OS versionNHI Mgmt Group practiceRestricted settings and newer protections need current Android releases

Where identity fits

Accessibility malware is a credential and token theft tool. It reads passwords as they are typed, one-time codes from SMS and authenticator apps, and wallet seed phrases. For workforce identity, that argues for phishing-resistant methods and short session lifetimes on mobile. For app owners, keep tokens and API keys out of anything shown on screen, and bind sessions to the device with attestation so stolen values are less useful elsewhere. See the Mobile App Attestation and Phishing-Resistant MFA Guide.

How to implement

  1. Inventory: report which accessibility services are enabled across managed Android devices and which apps provide them.
  2. Agree the allowlist: with your accessibility lead, list the assistive apps staff need so the policy does not block people who rely on them.
  3. Audit mode: alert on unapproved services for two to four weeks without blocking, and contact users of each finding.
  4. Enforce in rings: apply the allowlist and app source restrictions to a pilot ring, then widen. Roll back for an individual through a documented exception rather than lifting the policy.
  5. Connect response: route alerts to a playbook that isolates the device and revokes the user's sessions.

Decision guidance

  • Fully managed devices: enforce the allowlist and block unknown sources.
  • Work profile BYOD: the enterprise cannot see personal apps, so rely on MTD, conditional access and phishing-resistant MFA.
  • App owners: decide whether to warn, step up or block sensitive actions when an unrecognised accessibility service is active, balancing fraud loss against users who depend on assistive tools.

Responding to suspected accessibility abuse

  • Isolate: remove the device from corporate access through conditional access.
  • Revoke: end sessions, reset passwords from a clean device and re-register MFA, since codes and credentials may have been read.
  • Preserve: capture the app package name, MDM inventory and network logs before removal.
  • Remove: if the app blocks uninstall, use MDM removal or a factory reset after evidence is collected.
  • Notify: for banking or wallet exposure, tell the relevant institutions.

Detection and measurement

  • Unapproved services: managed devices with an unapproved accessibility service enabled (target zero).
  • Sideload exposure: devices with apps from unknown sources.
  • Code-based MFA on mobile: percentage of users relying on SMS or authenticator codes.
  • Time to isolate: from alert to conditional access block.
  • Allowlist exceptions: number and age of open exceptions.

Framework mapping

FrameworkReferenceRelevance
MITRE ATT&CK for MobileT1453; M1011; DET0697; related T1417.001, T1417.002, T1516Technique, mitigation, detection strategy and linked techniques
NIST SP 800-53 Rev. 5AC-19 Access Control for Mobile Devices; CM-7 Least Functionality; SI-4 System MonitoringRestricting device functions and monitoring
NIST CSF 2.0DE.CM-09; PR.AA-03Monitoring devices and software; authentication
OWASP MASVSMASVS-PLATFORMFor in-house apps: secure interaction with the platform

Common failure patterns

  • Allowing sideloading on corporate devices: the delivery route for most families in ATT&CK's list.
  • Blocking all accessibility services: staff who rely on assistive tools lose access and the policy gets lifted.
  • Trusting authenticator codes on the same phone: a service that reads the screen reads the code.
  • Treating Play distribution as proof of safety: ATT&CK's own user guidance warns otherwise.
  • Remediating without rotating credentials: removing the app leaves stolen passwords and sessions valid.

Practitioner checklist

  • Report enabled accessibility services across managed Android devices.
  • Agree an allowlist of assistive apps with your accessibility lead.
  • Restrict corporate devices to managed or official app stores.
  • Alert on unapproved accessibility services and on apps that request it soon after install.
  • Move mobile users from code-based MFA to phishing-resistant methods.
  • Teach users what an accessibility request looks like and where to check it.
  • Rotate credentials and sessions as part of every removal.

Standards and references

Related NHI Mgmt Group resources: MITRE ATT&CK for Mobile Guide · Android Banking Trojans Guide · Android Enterprise Security Guide · Mobile Malware Guide

PDF for subscribers

Take this guide with you

Every section, table and checklist in one PDF, with links to the sources and related guides. Subscribe to our weekly NHI & AI Identity Journal and we’ll email you a download link.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 12 November 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org

PDF Downloads for Newsletter Subscribers

Get the guide

Enter your details and we’ll email you a download link.

The full guide is free to read online. The PDF is a thank you for subscribing.

Fill in all fields marked * and tick the PDF box to continue.