prompts Archives

Winsage

September 30, 2026

Windows 11 setup now prompts you to install TikTok and Netflix apps during first boot

Microsoft has introduced a new feature called Assisted Setup in Windows 11, which allows users to install popular third-party applications like TikTok, WhatsApp, and Spotify immediately after the out-of-box experience (OOBE). This feature is gradually rolling out with build 26340.9354, released on September 8. Users can opt for automatic downloads and installations of these applications, but they must explicitly consent to download them, unlike pre-installed apps like OneDrive and Phone Link. The setup process also allows users to pin popular websites to the taskbar, with recommendations based on geographical location. Windows 11 26H2 has been officially released.

AppWizard

September 29, 2026

AI-powered app maker Wabi pivots to a messaging experience

Wabi, an AI startup, has launched Wabi 2.0, a platform that serves as a sophisticated AI messenger while maintaining app-building capabilities. The founder, Eugenia Kuyda, describes Wabi 2.0 as a "personal agent" that enhances user experience by integrating conversation and task management. The platform distinguishes itself by targeting AI agents with conversational and productivity features, rather than competing with traditional coding tools. Users can request tasks that the agent fulfills by generating a tailored user interface for specific needs, such as health tracking or family management. Access to Wabi 2.0 is currently limited to invite codes distributed via X.

AppWizard

September 29, 2026

How we found 24 Android vulnerabilities using our open source AI security agent

The GitHub Security Lab has introduced the Taskflow Agent, an open-source tool that uses artificial intelligence to automate the identification of vulnerabilities in Android applications. This tool allows security researchers to create custom taskflow prompts, leading to the discovery of over 20 vulnerabilities. To use the taskflows, a GitHub Copilot license is required, and users can run scripts to audit their projects, with results displayed in an SQLite viewer. Two specific vulnerabilities identified include: 1. OsmAnd app allows malicious applications to track user locations due to improper handling of intent extras in its exported activity. 2. The Wikipedia app has a logic flaw that enables attackers to redirect users to malicious sites via a deeplink mechanism, potentially leading to account takeovers. The GitHub Security Lab has reported a total of 24 vulnerabilities in Android applications, highlighting the effectiveness of AI in security research.

AppWizard

September 29, 2026

Gemini app adding Google Wallet integration for ‘personalized insights’

The Gemini Connected App has introduced a feature called Google Wallet, which provides users with tailored information based on their stored data, helping manage digital assets like loyalty cards and travel documents. Users can retrieve information such as boarding passes and airline loyalty numbers. The app also offers financial insights, including transaction data from services like Plaid, expense summaries, budgeting tips, and overviews of rewards linked to stored cards. This integration is currently exclusive to Google Wallet users in the U.S. and allows users to ask questions about their spending habits. However, users cannot execute transactions or modify linked financial accounts through the app. The feature is being gradually rolled out and is limited to personal Google Accounts in the U.S.

AppWizard

September 28, 2026

RatHat Android Malware Console Uses Gemini to Identify Higher-Value Victims

The operators behind RatHat have created an advanced Android banking trojan that allows control of infected devices through a web console. Since April 2026, nearly 100 instances of this console have been identified, indicating a malware-as-a-service model. The console collects sensitive data from compromised phones, including text messages and passwords. It uses Google's Gemini AI to assess victims' bank balances, categorizing them into high-value and mid-value segments, but does not facilitate financial transactions. The malware has remained largely unchanged since late 2025, but the console has seen significant updates, leading to three new versions: BlackCat Remote Control Management, Panda Workshop V5, and V6. These versions serve as both control interfaces and build tools for creating and distributing malware. The latest version includes templates for deceptive download pages. RatHat infiltrates devices via text messages and online ads, requesting Accessibility access to read screens and simulate user interactions. This access allows the malware to activate wireless debugging and connect to the Android Debug Bridge (ADB), giving operators elevated privileges to execute commands. A Go program can be deployed to maintain control, even after the app is uninstalled. Cleafy has tracked console deployments through web code analysis, noting that many IP addresses originate from a Singapore-registered network. The initial console version allowed operators to choose AI providers, but the latest exclusively uses Gemini. RatHat also utilizes Gemini on infected devices to determine tap locations based on screen layouts. Cleafy has compiled indicators related to the consoles' command-and-control servers, download links, and malware samples, including specific domains, IP addresses, and MD5 hashes. The consoles often use web addresses starting with "admin." and inexpensive top-level domains. Security tools are advised to monitor processes running under the shell user on affected devices.

AppWizard

September 28, 2026

I turned off these 8 Google Messages features, and the app is so much better now

Google Messages has evolved to include features such as read receipts, Gemini integration, typing indicators, and chat bubbles. Some users find these enhancements beneficial, while others view them as intrusive. The author disabled several features for a more streamlined experience, leading to less clutter and fewer distractions. The features disabled include: 1. Gemini Button: A floating icon for AI-assisted messaging, which can be disabled in Message settings. 2. Profile Sharing: Allows visibility of name and photo to contacts, with customizable privacy settings available. 3. Outgoing Message Sounds: Can be turned off to create a quieter messaging environment. 4. Floating Chat Heads (Bubbles): Can be disabled to avoid distractions and protect sensitive information. 5. Follow-Up Reminders: Prompts to respond to message threads, which can be turned off in Message settings. 6. Read Receipts and Typing Indicators: Create pressure to respond quickly and can be disabled in Message settings. 7. Sensitive Content Warnings: Can misidentify benign images and can be turned off for a smoother experience. 8. Smart Reply and Suggested Actions: Offers quick responses that may not fit all conversations and can be disabled. These adjustments have improved the author's messaging experience, allowing for more personal and less stressful interactions.

AppWizard

September 26, 2026

[Minecraft] Su Has a Talent For Getting Kidnapped

YouTube has introduced a passive sign-in feature that allows users to browse the platform without full authentication, enhancing user experience by simplifying access to content. This update is expected to increase viewer engagement and retention for content creators, potentially leading to greater monetization opportunities and audience growth. The new interface prompts users to log in only when necessary, aligning with current trends in user experience design.

AppWizard

September 25, 2026

F-Droid gets its biggest update in a decade with new UI and smoother app installs

F-Droid is a community-driven repository for free and open source Android applications, recently undergoing a transformation with the release of F-Droid 2.0. The new client is redesigned using Kotlin Compose, enhancing responsiveness and introducing optional Material theming for a visually appealing interface. The update simplifies app discovery and installation, now offering a vast library of thousands of open source applications. F-Droid prioritizes user autonomy without tracking behavior for ads. It features extensive categories for specialized needs and improved search functionality that returns results based on app names and descriptions. Installation has been streamlined using Google’s pre-approval API, allowing users to confirm installations through a simplified pop-up, improving the user experience despite requiring one additional click compared to the Play Store.

BetaBeacon

September 24, 2026

Move over, Claude and ChatGPT: This Android brand’s open-source AI model builds 3D games on demand

Xiaomi has open-sourced its omnimodal AI model, MiMo-V2.6 Pro, which can generate interactive 3D worlds from text or visual prompts. The model is claimed to be the most capable open-weight model on the Artificial Analysis Intelligence Index benchmark.

Winsage

September 23, 2026

Cisco Talos Discloses Autonomous Windows Malware: Four AI Models Direct Each Attack

Cisco Talos has disclosed a new Windows implant called CLOSEDQUORUM, which operates without human operators by using four commercial large language models (LLMs) to autonomously decide on actions after deployment. The implant is a 16.4-megabyte executable that conducts reconnaissance on the target system and sends structured prompts to the LLMs, which respond with one of four actions: steal, inject, persist, or move. When the "steal" action is chosen, it simultaneously attacks three credential stores, extracting Windows credentials, browser passwords, and cryptocurrency wallet data. The "inject" action uses either Early Bird APC injection or process hollowing based on model recommendations. For persistence, it employs three overlapping mechanisms, including a Registry Run key, a scheduled task, and a WMI event subscription. To evade detection, CLOSEDQUORUM suppresses ETW telemetry and introduces delays before executing actions. It circumvents traditional command-and-control structures by calling legitimate API endpoints, making blocking these domains impractical. Talos introduced the CAIRN toolkit for detecting AI-integrated malware, which operates on metadata and employs rule-based detection. Unlike previous AI-assisted malware, CLOSEDQUORUM automates decision-making processes entirely, highlighting a significant evolution in malware capabilities. Security teams are advised to focus on behavioral detection strategies and utilize the YARA rule and CAIRN toolkit for identifying this threat.