new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

Tauri · Develop · all subjects

security: permissions

9 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

Tauri 2.0 security audit

Security is a priority for Tauri 2.0. The project works with external auditors to review code for security. A similar security audit approach was used for Tauri 1.0 in collaboration with Radically Open Security.

readDir security vulnerability in Tauri 1.0.6 and 1.1.0

A security vulnerability was patched in Tauri 1.1.0 where the readDir function could return entries outside the configured scope when a symlink is found. The patch is also available in Tauri 1.0.6.

External API Access configuration for remote domain IPC

Tauri 1.3 adds dangerousRemoteUrlIpcAccess configuration to allow specified trusted (sub)domains to access the Tauri IPC layer. The configuration uses JSON format with properties: windows (array of window names), domain (trusted subdomain), plugins (array of plugin names), and enableTauriAPI (boolean). Example: {"security": {"dangerousRemoteUrlIpcAccess": [{"windows": ["main", "settings"], "domain": "trusted.example", "plugins": ["trusted-plugin"], "enableTauriAPI": false}]}}. Only specified (sub)domains are allowed; wildcards and glob patterns were rejected due to security risks. Shared domains must not be used. Domain ownership must not change over the application lifetime.

Browser arguments security restriction in Tauri 1.3

The additional_browser_args option for window creation was restricted to Rust-side only after security review. Frontend exposure was found to be highly risky as webview command-line arguments can enable dangerous features like profile stealing and security measure disabling. Custom Rust commands can be implemented for application-specific window creation with controlled arguments.

Tauri v2 replaces v1 allowlist with Access Control List permissions

Tauri v2 introduces a new Access Control List based approach for command access permissions, replacing the v1 allowlist. The new system provides fine-grained configuration with multiwindow support, allowing commands to be scoped to specific windows or remote URLs.

Core plugins namespace change - prepend with core:

Tauri 2.0 introduces a breaking change for core plugins. All core pseudo plugins now require the 'core:' namespace prefix in capabilities permissions. Previous format like 'path:default' must be changed to 'core:path:default'. Alternatively, use the new 'core:default' permission set which contains all default permissions of all core plugins.

Core plugins list requiring namespace change

Core pseudo plugins that require 'core:' prefix in permissions are: path, event, window, app, image, resources, menu, and tray.

core:default permission scope

The 'core:default' permission set grants all default permissions of all core plugins and is considered reasonably secure and safe to enable by default, with limited impact in case of a compromised frontend.

Security model: permissions, scopes, and capabilities

Tauri 2.0 uses a new access control system replacing the allowlist. The system consists of permissions (on-off toggles for Tauri commands), scopes (parameter validation for Tauri commands), and capabilities (attaching permissions and scopes to Windows and WebViews). Named permission or scoping files can be created and reused in combination with other named permissions or scopes to build fine-grained access control. Plugin developers can define default permissions based on secure-by-default security assumptions. Application developers can use, extend, or reduce plugin permissions.

Give your agent this brain