This is an archived page. It's not actively maintained.

Firefox OS security overview

This document provides an overview of Mozilla's Firefox OS security framework, which is designed to protect mobile devices from threats to the platform, apps, and data. In Firefox OS, Mozilla has implemented a comprehensive, integrated, and multi-layered security model that delivers best-of-breed protection against security risks to mobile phones.

Platform Security

The Firefox OS platform uses a multi-layered security model that is designed to mitigate exploitation risks at every level. Front-line countermeasures are combined with a defense-in-depth strategy that provides comprehensive protection against threats.

Secure Architecture

The Firefox OS connects web-based applications to the underlying hardware. It is an integrated technology stack consisting of the following levels:

  • Gaia: The suite of web apps that make up the user experience (apps consist of HTML5, CSS, JavaScript, images, media, and so on).
  • Gecko: The application runtime layer that provides the framework for app execution, and implements the Web APIs used to access features in the mobile device.
  • Gonk: The underlying Linux kernel, system libraries, firmware, and device drivers that everything runs on top of.
  • The mobile device: The mobile phone running Firefox OS.

Gecko is the gatekeeper that enforces security policies designed to protect the mobile device from misuse. The Gecko layer acts as the intermediary between web apps (at the Gaia layer) and the phone. Gonk delivers features of the underlying mobile phone hardware directly to the Gecko layer. Web apps access mobile phone functionality only through the Web APIs, and only if Gecko allows the access request — there is no direct access, no “back door” into the phone. Gecko enforces permissions and prevents access to unauthorized requests.

Secure System Deployment

Firefox OS comes installed on a smart phone. The original system image is created by a known, trusted source — usually the device OEM — that is responsible for assembling, building, testing, and digitally signing the distribution package.

Security measures are used throughout the technology stack. File system privileges are enforced by Linux's access control lists (ACLs). System apps are installed on a volume that is read-only (except during updates, when it is temporarily read-write); generally only areas containing user content may be read-write. Various components within the device hardware have built-in protections that are implemented by default as standard industry practice — chipset manufacturers, for example, employ hardening techniques to reduce vulnerabilities. The core platform (Gecko and Gonk) is hardened to strengthen its defense against potential threats, and hardening features of the compiler are used where applicable. For further details see Runtime security.

Secure System Updates

Subsequent upgrades and patches to the Firefox OS platform are deployed using a secure Mozilla process that ensures the ongoing integrity of the system image on the mobile phone. The update is created by a known, trusted source — usually the device OEM — that is responsible for assembling, building, testing, and digitally signing the update package.

System updates can involve all or a portion of the Firefox OS stack. If changes to Gonk are included in the update, then FOTA (Firmware Over the Air) is the install process used. FOTA updates can also include any other part of the Firefox OS stack, including device management (FOTA, firmware / drivers), settings management (Firefox OS settings), security updates, Gaia, Gecko, and other patches.

Updates that do not involve Gonk can be done using the Mozilla System Update Utility. Firefox OS uses the same update framework, processes, and Mozilla ARchive (MAR) format (used for update packages) as the Firefox Desktop product.

A built-in update service — which may be provided by the OEM — on the mobile phone periodically checks for system updates. Once a system package becomes available and is detected by the update service, the user is prompted to confirm installation. Before updates are installed on the mobile device, the device storage is checked for sufficient space to apply the update, and the distribution is verified for:

  • Update origin (verify the source location protocol:domain:port of the system update and manifest)
  • File integrity (SHA-256 hash check)
  • Code signature (certificate check against a trusted root)

The following security measures are used during the update process:

  • Mozilla recommends and expects that updates are fetched over an SSL connection.
  • Strong cryptographic verification is required before installing a firmware package.
  • The complete update must be downloaded in a specific and secure location before the update process begins.
  • The system must be in a secure state when the update process starts, with no web apps running.
  • The keys must be stored in a secure location on the device.

Rigorous checks are in place to ensure that the update is applied properly to the mobile phone.

Note: For more information on how thee updates work and how to create and distribute updates, read Creating and applying Firefox OS update packages.

App Security

Firefox OS uses a defense-in-depth security strategy to protect the mobile phone from intrusive or malicious applications. This strategy employs a variety of mechanisms, including implicit permission levels based on an app trust model, sandboxed execution at run time, API-only access to the underlying mobile phone hardware, a robust permissions model, and secure installation and update processes. For technical details, refer to Application security.

In Firefox OS, all applications are web apps — programs written using HTML5, JavaScript, CSS, media, and other open web technologies (pages running within the browser are not referred to as web apps in this context). Because there are no binary ("native") applications installed by the user, all system access is mediated strictly through the Web APIs. Even access to the file system happens only through Web APIs and a back-end SQLite database — there is no direct access from apps to files stored on the SD card.

Firefox OS limits and enforces the scope of resources that can be accessed or used by an app, while also supporting a wide range of apps with varying permission levels. Mozilla has implemented tight control over what type of applications can access which APIs. For example, only certified apps (shipped with the phone) can have access to the Telephony API. The Dialer app has privileges to access the Telephony API in order to make phone calls, but not all certified apps can access this API.

This prevents a situation, for example, in which an arbitrary third-party app gets installed, dials a pay-per-use phone number (900 and 910), and racks up a large cell phone bill.

Other OEM apps might be selectively given access to the Telephony API, however. For example, an operator might provide a systems management application that allows a customer to manage their account, including the ability to phone the Operator’s billing or support office directly.

Trusted and Untrusted Apps

Firefox OS categorizes apps according to the following types:


Trust Level



Highly Trusted

System apps that have been approved by the Operator or OEM (due to risk of device corruption or risk to critical functionality). System apps and services only; not intended for third-party applications.
This designation is reserved for just a small number of critical applications. Examples: SMS, Bluetooth, camera, system clock, telephony, and the default dialer (to ensure that emergency services are always accessible).



Third-party apps that have been reviewed, approved, and digitally signed by an authorized Marketplace.

Web (everything else)


Regular web content. Includes both installed apps (stored on the mobile phone) and hosted apps (stored remotely, with only an app manifest stored on the mobile phone). The manifest for hosted apps can be obtained through a Marketplace.

An application’s trust level determines, in part, its ability to access mobile phone functionality.

  • Certified apps have permissions to most Web API operations.
  • Privileged apps have permissions to a subset of the Web API operations accessible to Certified apps.
  • Untrusted apps have permissions to a subset of the Web API operations accessible to Privileged apps — only those Web APIs that contain sufficient security mitigations to be exposed to untrusted web content.

Some operations, such as network access, are assumed to be an implicit permission for all apps. In general, the more sensitive the operation (for example, dialing a phone number or accessing the Contacts list), the higher the app trust level required to execute it.

Note: for more information on the APIs available and their permission levels, consult App permissions.

Principle of Least Permissions

For web apps, the Firefox OS security framework follows the principle of least permissions: start with the absolute minimum permissions, then selectively grant additional privileges only when required and reasonable. By default, an app starts with very low permissions, which is comparable to untrusted web content. If the app makes Web API calls that require additional permissions, it must enumerate these additional permissions in its manifest (described later in this document). Gecko will consider granting Web API access to an application only if the applicable privileges are explicitly requested in its manifest. Gecko will grant the requested permission only if the type of the Web App (certified, trusted, or web) is sufficiently qualified for access.

Review Process for Privileged Apps in the Marketplace

In order for an app to become privileged, the app provider must submit it for consideration to an authorized Marketplace. The Marketplace subjects the app to a rigorous code review process: verifying its authenticity and integrity, ensuring that requested permissions are used for the purposes stated (in the permission rationale), verifying that the use of implicit permissions is appropriate, and validating that any interfaces between privileged app content and unprivileged external content have the appropriate mitigations to prevent elevation of privilege attacks. The Marketplace has the responsibility to ensure that the web app will not behave maliciously with the permissions that it is granted.

After an app passes this review, it is approved for use, its app manifest is digitally signed by the Marketplace, and it is made available for mobile users to download. The signature ensures that, if the web store were somehow hacked, the hacker could not get away with installing arbitrary content or malicious code on users’ phones. Due to this vetting process, Firefox OS gives privileged apps obtained from a Marketplace a higher degree of trust than everyday (untrusted) web content.

Note: to find out more about Marketplaces including the Firefox Marketplace, go to the Marketplace zone.

Packaged and Hosted Apps

Apps for Firefox OS can be either packaged (stored on the mobile phone) or hosted (stored on a remote web server, with just a manifest stored on the mobile phone). There are some differences in the way in which security is managed for each. Nonetheless, packaged and hosted apps are both subject to application sandboxing, which is described later in this document.

Note: You can find out more about hosted and packaged apps at App publishing options.

Packaged Apps

A packaged app consists of a ZIP file containing application resources (HTML5, CSS, JavaScript, images, media), as well as a manifest that provides an explicit list of assets and their corresponding hashes. Certified and privileged apps must be packaged apps because the app manifest needs to be digitally signed. When a user obtains a packaged app, the ZIP file is downloaded onto the mobile phone, and the manifest is read from a known location inside the ZIP file. During the install process, app assets are verified and remain stored locally in the package. All explicit permissions are requested at runtime, showing the user the app's data usage intentions, and persisted by default.

To refer to app resources in a packaged app, the URL begins with app: using the following format:


where app:// represents the mount point for the ZIP file, and identifier is a UUID that is generated when the app is installed on the mobile phone. This mechanism ensures that resources referred to with an app: URL are contained in the ZIP file. The path within an app: is relative, so relative links to resources in the ZIP file are allowed.

While packaged apps are primarily intended to be used for Certified or Privileged apps, regular web apps can also be packaged. However, they do not gain any increase in trust or permissions access simply because they are packaged.

Hosted Apps

Hosted apps are located on a web server and loaded via HTTP. Only the app manifest is stored on the mobile phone. Everything else is stored remotely. Certain APIs are available only to privileged and certified apps, which requires the app to be packaged due to signing requirements. Therefore, a hosted app will not have access to any of the Web API operations that require privileged or certified app status.

From a security point of view, hosted apps work very much like normal websites. A hosted app is loaded by invoking a hard-coded, fully-qualified URL that points to the startup page in the root directory of the app on that web server. Once a hosted app is loaded, the mobile phone links to pages using the same URLs that are used when browsing the web site.

App Manifest

An Open Web App manifest contains information that a Web browser needs in order to interact with an app. A manifest is a JSON file with (at a minimum) a name and description for the app. For further details, refer to FAQs about app manifests.

Example Manifest

The following code listing shows an example manifest with just basic settings:

  "name": "My App",
  "description": "My elevator pitch goes here",
  "launch_path": "/",
  "icons": {
    "128": "/img/icon-128.png"
  "developer": {
    "name": "Your name or organization",
    "url": ""
  "default_locale": "en"

Security Settings in the App Manifest

The manifest can also contain other settings, including the following security settings:




Permissions required by the app. An app must list every Web API it intends to use that requires user permission. Most permissions make sense for privileged apps or certified apps, but not for hosted apps. Properties per API:

  • description: A string specifying the intent behind requesting use of this API. Required.
  • access: A string specifying the type of access required for the permission. Implicit permissions are granted at install time. Required for only a few APIs. Accepted values: read, readwrite, readcreate, and createonly.


The Origin of the app; can be singular or an array of origins (scheme+unique hostname) that are allowed to trigger installation of this app. Allows app providers to restrict installs from only an authorized Marketplace (such as


Content Security Policy (CSP). Applied to all pages loaded in the app. Used to harden the app against bugs that would allow an attacker to inject code into the app. If unspecified, privileged and certified apps have system-defined defaults. Syntax:

Note that this directive can only increase the CSP applied. You cannot use it, for example, to reduce the CSP applied to a privileged App.


Type of application (web, privileged, or certified).

Firefox OS requires that the manifest be served with a specific mime-type (application/x-web-app-manifest+json) and from the same fully-qualified host name (origin) from which the app is served. This restriction is relaxed when the manifest app (and thus the app manifest) is same-origin with the page that requested the app to be installed. This mechanism is used to ensure that it's not possible to trick a website into hosting an application manifest.

Sandboxed Execution

This section describes application and execution sandboxes.

Application Sandbox

The Firefox OS security framework uses sandboxing as a defense-in-depth strategy to mitigate risks and protect the mobile phone, platform, and data. Sandboxing is a way of putting boundaries and restrictions around an app during run-time execution. Each app runs in its own worker space and it has access only to the Web APIs and the data it is permitted to access, as well as the resources associated with that worker space (IndexedDB databases, cookies, offline storage, and so on).

The following figure provides an overview of this security model.

By isolating each app, its impact is contained within its own worker space and it cannot interfere with anything (such as other apps or their data) outside of that worker space.

Execution Sandbox

B2G (Gecko) runs in a highly-privileged system process that has access to hardware features in the mobile phone. At runtime, each app runs inside an execution environment that is a child process of the B2G system process. Each child process has a restricted set of OS privileges — for example, a child process cannot directly read or write arbitrary files on the file system. Privileged access is provided through Web APIs, which are mediated by the parent B2G process. The parent ensures that, when a child process requests a privileged API, it has the necessary permission to perform this action.

Apps communicate only with the B2G core process, not with other processes or apps. Apps do not run independently of B2G, nor can apps open each other. The only “communication” between apps is indirect (for example, when one app sets a system alarm and another app triggers a system notification as a result of it), and is mediated through the B2G process.

Hardware Access Only via the Web API

Web apps have only one entry point to access mobile phone functionality: the Firefox OS Web APIs, which are implemented in Gecko. Gecko provides the sole gateway to the mobile device and underlying services. The only way to access device hardware functionality is to make a Web API call. There is no “native” API and there are no other routes (no “back doors”) to bypass this mechanism and interact directly with the hardware or penetrate into low-level software layer.

Security Infrastructure

The following figure shows the components of the Firefox OS security framework:

  • Permission Manager: Gateway to accessing functionality in the Web API, which is the only access to the underlying hardware.
  • Access Control List: Matrix of roles and permissions required to access Web API functionality.
  • Credential Validation: Authentication of apps/users.
  • Permissions Store: Set of privileges required to access Web API functionality.

Permissions Management and Enforcement

Firefox OS security is designed to verify and enforce the permissions granted to web apps.

The system grants a particular permission to an app only if the content requests it, and only if it has the appropriate permissions requested in the app’s manifest. Some permissions require further authorization from the user, who is prompted to grant permission (as in the case of an app requesting access to the user’s current location). This app-centric framework provides more granular control over permissions than traditional role-centric approaches (in which individual roles are each assigned a set of permissions).

A given Web API has a set of actions and listeners. Each Web API has a required level of permission. Every time a Web API is called, Gecko checks permission requirements (role lookup) based on:

  • Permissions associated with calling app (as specified in the manifest and based on the app type.)
  • Permissions required to execute the requested operation (Web API call.)

If the request does not meet the permission criteria, then Gecko rejects the request. For example, untrusted apps cannot execute any Web APIs that are reserved for trusted apps.

Prompting Users for Permission

In addition to permissions that are implicitly associated with the web apps, certain operations require explicit permission from the user before they can be executed (for example, "can the web app access your camera?"). For these operations, web apps are required to specify, in their manifest, their justification for requiring this permission. This data usage intention informs users about what the web app intends to do with this data if permission is granted, as well as any risk involved. This allows users to make informed decisions and maintain control over their data.

Secure App Update Process

For app upgrades and patches to a privileged app, app providers submit the updated package to an authorized Marketplace, where it is reviewed and, if approved, signed and made available to users. On Firefox OS devices, an App Update Utility periodically checks for app updates. If an update is available, then the user is asked whether they want to install it. Before a update is installed on the mobile device, the package is verified for:

  • Update origin (verify the source location protocol:domain:port of the update and manifest)
  • File integrity (SHA-256 hash check)
  • Code signature (certificate check against a trusted root)

Rigorous checks are in place to ensure that the update is applied properly to the mobile phone. The complete update package must be downloaded in a specific and secure location before the update process begins. Installation does not overwrite any user data.

Note: For more information on app updates, read Updating apps.

Device Security (Hardware)

Security mechanisms for the mobile device hardware are typically handled by the OEM. For example, an OEM might offer SIM (Subscriber Identity Module) card locks, along with PUK (PIN Unlock Key) codes to unlock SIM cards that have become locked following incorrect PIN entries. Contact your OEM for details. Firefox OS does allow users to configure passcodes and timeout screens, which are described in the next section.

Data Security

Users can store personal data on their phone that they want to keep private, including contacts, financial information (bank & credit card details), passwords, calendars, and so on. Firefox OS is designed to protect against malicious apps that could steal, exploit, or destroy sensitive data.

Passcode and Timeout Screens

Firefox OS allows users to set a passcode to their mobile phone so only those who supply the passcode can use the phone. Firefox OS also provides a timeout screen that is displayed after a configurable period of phone inactivity, requiring passcode authentication before resuming use of the phone.

Sandboxed Data

As described earlier, apps are sandboxed at runtime. This prevents apps from accessing data that belongs to other apps unless that data is explicitly shared, and the app has sufficient permissions to access it.

Serialized Data

Web apps do not have direct read and write access to the file system. Instead, all access to storage occurs only through Web APIs. Web APIs read from, and write to, storage via an intermediary SQLite database. There is no direct I/O access. Each app has its own data store, which is serialized to disk by the database.

Data Destruction

When a user uninstalls an app, all of the data (cookies, localStorage, IndexedDB, and so on) associated with that application is deleted.


Mozilla is committed to protecting user privacy and user data according to its privacy principles (, which stem from the Mozilla Manifesto ( The Mozilla Firefox Privacy Policy describes how Mozilla collects and uses information about users of the Mozilla Firefox web browser, including what Firefox sends to websites, what Mozilla does to secure data, Mozilla data practices, and so on. For more information, see:

Firefox OS implements these principles by putting the control of the user data in the hands of the user, who gets to decide where this personal information goes. Firefox OS provides the following features:

  • Do Not Track option
  • ability to disable Firefox browser cookies
  • ability to delete the Firefox OS browsing history