Cordial Homes left behind unhappy customers and contractors. Now, its “silver tongued” director has left the country, and his former business partner has been bankrupted. Read...

Official support for Windows 10 was supposed to end in October 2025, but Microsoft shifted plans and continued to provide security updates for another year via the Extended Security Updates (ESU) program. It’s free for private users and protects against malware and other security threats until at least October 2026. Businesses, on the other hand, can keep getting security updates for a fee for three years. In a recent Windows IT Pro blog post, Microsoft said that it will extend the ESU program to include additional Windows versions. This means that other operating systems that are approaching the end of their support will receive updates for longer than originally planned. Which additional Windows versions? Windows 10 Enterprise LTSB 2016, Windows 10 IoT Enterprise 2016 LTSB, and Windows Server 2016. The lifecycles of the first two were supposed to end on October 13th, 2026, but will now receive updates for three years longer. As for Windows Server 2016, it was scheduled to end support in January 2027 but will also receive support for another three years. Reasons, costs, and details Participation in the ESU program is voluntary and subject to a fee, as these are primarily versions for businesses. As with the extended security updates for Windows 10 22H2, continued use for business customers is associated with increased costs year over year. In the first year, it will cost $61 per device. Afterwards, it will double in price every subsequent year: $122 for the second year and $244 for the third year. However, corporations will receive a discount if their devices are managed via Microsoft’s own cloud solutions (Intune or Autopatch). In this case, the first year costs only $45. The offer is aimed explicitly at companies and servers that are unable to upgrade to a newer version immediately. These upgrades are often costly, complex, and time-consuming to make. In addition, certain systems that run on Windows must remain stable and reliable—outages due to sudden updates are often not an option. Not a long-term solution Microsoft emphasizes that the ESU program is not a long-term solution, even for businesses. Although the three-year grace period is useful for continuing to receive security updates, the extended security updates do not include new features. Get Windows 11 Pro for cheap Windows 11 Pro Microsoft therefore recommends switching to the latest versions (either Windows 11 25H2, Windows Server 2025, or Windows Enterprise LTSC 2024) as soon as possible. Depending on the size of a company, the ongoing license costs for the ESU program may well be more expensive than a complete switch to a newer version. For companies that continue into the second year of the ESU, they’ll have to pay $122 per device. Read...

Scams keep coming at us—and they’re getting harder to spot. How? Scammers have begun making them more tailored to their marks. That is… us. Personalized scams, as security experts call them, use details about you in the hope of tricking you more easily. This information comes from illicit sources like data leaks and breaches, successful phishing attacks, compromised websites, and malware, as well as legitimate sources like marketing info, public records, and social media. As you might guess, the data range can span a pretty wide range, from location to shopping habits.  But what does a personalized scam look like? And how do you spot one? I brought these questions (and others) to a chat with Steve Grobman, Chief Technology Officer at McAfee—and it turns out that just like the data a scammer might have on you, the types of scams they craft fall under pretty broad umbrellas, too. The “general” personalized scam PCWorld These kinds of scams tend to target broad groups—like a specific geographic area. Toll scams have become personalized, for example. Before, messages claiming you had unpaid toll charges were generic. Now the texts will refer to your area’s toll authority and the name of the system, based on your phone number’s area code. If you’re not naturally a suspicious person, this updated approach may catch you off-guard. The language sounds more natural, despite being very broad. What changed? AI. Scammers can use AI to figure out regional information and incorporate it into messages quickly.  Scammers don’t have to know much about you to make this kind of connection. They’ll extrapolate it from your contact info. Think area code for a phone number or a specific service related to your email provider. For example, I’ve recently seen emails related to Google storage limits, claiming your files will be deleted soon because you ran out of space. The “specific” personalized scam Here’s where all those data leaks and breaches become a problem. Even when a data dump only involves details like name and location (like, say, from an address), a resulting scam message can sound much more official. It can address you by name, target your age bracket, and/or zero in on something specific to your region. The extra information allows for additional customization of the message. Grobman calls these “fill in the blank” scams, where a scary notice can easily swap in your name and a relevant entity to spook you. For my location, he described it as “___(name)____, the California Department of _________.” (A possible example would be: John, the California Department of Motor Vehicles has revoked your registration due to unpaid fees.) If matched well enough to your region, this approach could get you to click or otherwise fall for the scam, because it sounds realistic enough. The “hyperpersonalized” scam Scammers can target topical interests for their attacks.Wegovy This type of personalized scam is more insidious than outright creepy. (Mostly.) Grobman says these are “lifestyle” focused. Scammers use what they know of your habits—like sites you’ve visited or links you’ve clicked on—to figure out your interests. Then they’ll zero in to exploit that info. For example, if you’ve shown interest in weight loss, you could be targeted with a link to a fake weight loss drug. Hyperpersonalized scams can also take longer to build to the fraud—think romance scams, where the scammer uses information about you to build trust. Maybe they know where you went to school, and use that to start and build rapport. The more you share, the more they weave that into the relationship being built. Eventually, the requests for favors and money begin. Or shared communication, photos, and other details are then twisted into blackmail material, used to extort money in exchange for secrecy. Often, these kinds of scams can feel so personal—and so shameful to have fallen for one—that many victims won’t tell anyone they’ve been scammed. Previously, the young and the elderly were bigger targets for scammers, as they could prey on not just loneliness, but also lack of experience or diminished cognitive capabilities. But now, the threat for this to be widespread across all age brackets hangs lower than before… which is why we have to be on alert. What to do if you’ve been scammed First, take a deep breath. You might feel overwhelmed by your feelings—whether that’s shock, embarrassment, or shame—but that’s common and normal. Also common and normal: Making this kind of mistake. Falling for a scam can truly happen to anyone, even seasoned security professionals. Next, ask for help. The problem may seem huge at first, but getting help keeps the problem from spiraling into a huge mess. You can start with the FTC’s consumer advice page, which lists common scam scenarios and what steps to take afterward. Generally, you want to address the immediate problem first. Let’s say you used your credit card number on a scam site or wired money to a “special friend”—alert your bank about these fraudulent transactions right away. The faster you act, the faster you limit the damage. Worried about credit card or bank account fraud? Call your bank immediately!Cardmapr / Unsplash Or you shared your social security number and then realized your mistake. Add a security freeze to your credit reports immediately, and also add a security alert for good measure. (The freeze is the more powerful tool though, as it blocks anyone from checking or opening credit in your name until you temporarily allow access, aka “thaw” your report.) Take care of your emotional health, too. At a baseline, talking to a friend or family member who can help provide clarity or good feedback can help while you’re in a stew. You can also try your employer or even the police if you just need help in getting oriented. How to avoid personalized scams The grim reality is that personalized scams could become more common—the tools to help fraudsters keep improving, thanks to AI advances. (Thanks but no thanks, AI.) How fast that will happen remains to be seen—Grobman says scammers are business owners. They do what makes money, so a change in approach only comes when current efforts lose profitability. And that will happen the more awareness spreads and detection tools improve. (Remember, security experts also have AI available to them, too.) As that race continues to escalate, the shift toward more and more personalization will increase the difficulty of spotting legitimate messages among the fakes. Windows Security is a solid free antivirus option that Microsoft automatically keeps up to date.Foundry Fortunately, the best steps to protect yourself are also the easiest. Have antivirus software active on your PC. Be wary about installing apps on your computer or phone. Keep your software up to date, especially your browser. Use a password manager. Apps and services have begun to build in more safeguards and protections—think of it as a neighborhood watch approach to online security. The final piece of the puzzle? You. The sites you choose to visit, the software you download, the browser extensions you install, the links you click in email and messages—those all can increase or decrease your risk of getting caught in a scam, too. Surf the internet wisely. Read...

Windows 11 26H2 follows the model of an enablement package and is technically based on version 25H2. An enablement package is a small feature update that does not trigger a complete reinstallation of the operating system. Microsoft uses this process to unlock existing but previously disabled features that were added to the system via regular monthly updates. The enablement package essentially changes the version identifier of Windows and activates these features collectively without comprehensively replacing system files or affecting existing applications and settings. The installation process takes only a few minutes and is technically more like a cumulative update than a classic feature upgrade. The update to 26H2 changes the version identifier without reinstallation and extends the support period by 24 months for consumer editions and 36 months for business editions. Functional changes already reach systems via monthly updates for 25H2 and then carry over unchanged to 26H2. A significant increase in functionality is particularly evident on systems migrating from older versions such as 23H2. The rollout is scheduled for the familiar time window from late September to early October 2026. Those participating in the Insider Program will, of course, receive the new features in advance as a test version. Save 60% on Windows 11 Pro Windows 11 Pro Copilot-based search on the taskbar The classic Windows search is being replaced by an optional “Ask Copilot” feature. This can be activated in the taskbar settings under “Personalization.” Once activated, a compact search window opens with direct access to applications, files, and system settings. The technical basis remains the existing Windows search index. Thomas Joos Copilot also interprets the user’s intentions and links search queries to appropriate settings, such as brightness or display parameters. The interface does not display Bing pop-ups and responds noticeably more directly. A later extension will include file and image uploads as well as Copilot Vision, which incorporates content from open applications. Without activation, the familiar Windows search remains unchanged on the taskbar. Copilot does not access local data unless files are explicitly transferred. File Explorer with revised context structure File Explorer has been given a restructured context logic with nested menus to reduce the number of visible entries. Actions such as compression, path copying, or image rotation are grouped under “Manage file.” Image-related functions such as background assignment or rotation are also grouped together. Compression formats can be selected in a separate submenu. OneDrive-related options are moved to a separate area. Individual commands such as “Ask Copilot,” “Edit with Clipchamp,” “Edit in Notepad,” “Edit in Paint,” and Microsoft Photos still appear multiple times. Context-dependent prioritization via ML does not take place. At the same time, Microsoft is working on a dark display for the Properties tab in File Explorer. In addition, background preloading of Explorer is being tested to speed up startup, which can be disabled in the folder options. Copilot integration in File Explorer In addition, there will be deeper Copilot integration in File Explorer. The plan is to have a dockable side panel near the detail and preview views with the option to detach it into a separate window. Unlike the current forwarding of files to the Copilot app, Copilot operates directly in the file system and allows dialog-based interactions with folders and files. Thomas Joos The Run window, which has remained unchanged for decades, will get a WinUI-based variant with a Mica background, an enlarged input field, and a command overview above the input. It can be activated in the advanced section of the system settings. Once activated, the system will hide the classic variant. The new interface displays matching applications, including icons, as you type. At the same time, Microsoft is testing a dark mode for the classic Run window. Both versions coexist as options. Return of the agenda in the notification center The agenda view returns to the notification center and integrates with Outlook. After logging in, appointments are automatically synchronized and appear in the Win+N calendar view. The implementation uses WebView2 and occupies over 100MB of RAM. The view allows you to join meetings and view current appointments in real time. A Copilot button complements the interface. System Monitor as an integrated security component Sysinternals System Monitor is integrated directly into Windows 11. The component comes from the external Sysinternals suite and logs system events for threat detection. Events are stored in the Windows event database and can be filtered using separate configuration files. Activation is optional via Windows features or the command line. No restart is required. Camera settings and emoji update The camera settings have been expanded to include pan and tilt control for supported hardware. The options appear in the settings under “Bluetooth and devices” in the basic camera settings. The interface currently still shows inconsistencies in window adjustment. At the same time, Microsoft is introducing Emoji version 16.0 with new symbols. Gaming mode with Xbox full-screen interface An optional full-screen mode replaces the desktop interface with a controller-oriented dashboard based on the existing Xbox apps. It can be activated in the gaming settings. The mode reduces background processes and saves system resources, which has an impact on gaming performance and battery life. A restart activates the new interface. You can switch back to the desktop using the Windows key. For productive work, it is recommended to deactivate it before restarting. Agentic AI functions in the system An experimental area for agentic functions appears in the AI components of the settings. Applications such as Copilot take on automated tasks related to file organization, scheduling, or emailing. The function remains disabled until you turn it on and use it. Thomas Joos Bug fixes and detail improvements Current Windows builds also fix visual glitches with the automatically hidden taskbar, the unexpected rebuilding of desktop icons, and display issues with Windows security login windows. File Explorer adjustments improve accessibility, allow custom folder names, and add tooltips to favorites. A fix addresses application hangs during file operations with OneDrive and Dropbox, as well as problems with Outlook configurations that store PST files in OneDrive. Voice Access extends language support to include the Netherlands. Read...

It reflects `a huge amount of hard work`, supported by easing inflation and improving business and consumer confidence, chief executive Michael Boggs says. Read...

If you’ve been using your flat-rate Claude or Gemini account to feed OpenClaw and its eye-popping AI abilities, get ready to be banned. Specifically, those who’ve used their Claude and Google OAuth credentials for OpenClaw, the viral AI sensation that can gobble up millions of AI tokens in a single afternoon, have seen both Anthropic and Google ban their AI accounts. The bans frequently come without warning, and many who’ve tried to find out what happened or get their accounts reinstated have been met with silence. Of course, you can always dodge the banhammer by creating a new Claude or Google account, and it’s worth noting that most banned Google AI users are still able to access their Gmail, Google Drive, and other core Google services. Still, it’s not fun to get locked out of the $200-a-month Claude or $250-a-month Google AI account you’ve been actively using for OpenClaw and other tools, including Google’s popular Antigravity coding tool, particularly given the lack of refunds. “Pretty draconian from Google,” wrote OpenClaw creator Peter Steinberger on X, who added that he would likely remove support for using Google’s Antigravity OAuth credentials to power the viral AI agent. “Be careful out there if you use Antigravity.” Responding to the hubbub on X, Google DeepMind engineer Varun Mohan said the company has “been seeing a massive increase in malicious usage of the Anitgravity backend that has tremendously degraded the quality of service for our users. We needed to find a path to quickly shut off access to these users that are not using the product as intended. We understand that a subset of these users were not aware that this was against our ToS and will get a path for them to come back on but we have limited capacity and want to be fair to our actual users.” While both Anthropic and Google are cracking down on the use of flat-rate OAuth credentials for OpenClaw, ChatGPT isn’t wielding the banhammer yet–perhaps because OpenClaw creator Steiberger is now among OpenAI’s newest employees. OpenClaw itself is still an open-source tool, albeit with the backing of OpenAI. To understand the hubbub over Anthropic and Google’s bans for OpenClaw use via OAuth credentials, you must know the difference between OAuth and API access to Claude and Gemini. Even if you don’t know what OAuth access is, you likely use it all the time. Whenever you log into a third-party service using one of those “Login with Google,” “Login with Facebook,” or “Login with Apple” buttons, that’s OAuth at work. Now, the Claude and Google bans for OAuth users connecting to OpenClaw aren’t about OAuth per se; instead, it’s the fact that OAuth credentials are used for authenticating flat-rate Claude and Gemini plans. Generally speaking, you’re not supposed to use your Claude or Google AI OAuth credentials to power third-party AI tools and services, which typically don’t enforce rate-limiting policies; instead, you’re supposed to use the Claude or Gemini API, which charge by the token rather than a flat monthly fee. (In AI lingo, a token is a tiny unit of data used by AI models to process your prompts and create content such as text, images, and video; the more you interact with an AI, the more tokens you burn.) Still, just because you’re not supposed to use your Claude, ChatGPT, or Google OAuth credentials for an unsanctioned third-party service doesn’t mean you can’t. For OpenClaw users to authenticate their flat-rate Claude accounts with the bot, they can grab an OAuth token (not to be confused with an LLM token) by running the Claude Code setup process, then using that token for OpenClaw rather than Claude Code. For Gemini users, a Google Antigravity plug-in for OpenClaw does the trick. (The process for connecting ChatGPT to OpenClaw via OAuth is far easier, but remember, OpenAI just hired OpenClaw’s creator and is thus more invested in OpenClaw’s success.) But while big AI providers like Anthropic and Google may have tolerated such “hacky” OAuth authentication methods in the past, they’re now cracking down in earnest, and the explosive popularity of OpenClaw has accelerated the process. Unlike previous third-party AI tools that more or less flew under the radar, OpenClaw went from obscurity to everywhere-all-at-once status in a matter of weeks, far outstripping the user base of previous tools.  Another difference is that OpenClaw’s jaw-dropping agentic AI abilities come at the expense of tokens–a lot of tokens, with some users astonished to find they’d used millions of tokens in a single afternoon. Simply asking your OpenClaw agent “how are you?” could chew through 30,000 tokens (or way more, if you have a lengthy OpenClaw session going), versus a couple thousand tokens for ChatGPT on the web. From a business perspective, Anthropic and Google probably saw money going down the drain as OpenClaw users let it rip with their OAuth-enabled flat-rate AI accounts. And to the point of Google DeepMind’s Varun Mohan, the OpenClaw-driven surge in OAuth usage may have taken a toll on non-OpenClaw users on the same flat-rate plans. (Indeed, I noticed frequent “attempting to reach Gemini 3 Flash” warnings when vibe coding with Gemini CLI over the weekend. Were those connectivity issues were due to OpenClaw usage? Good question.) To repeat, Anthropic and Google are more than happy to fund your OpenClaw habit via a pay-as-you-go API key, and indeed, some API-using OpenClaw users have run up some eye-watering bills. But as far as using your flat-rate, OAuth-enabled Claude or Google account for OpenClaw, the party appears to be over. Read...

Wellingtonian Jess Carter has been self-employed as a digital illustrator for eight years, but with the onset of AI, her client demand has dropped more than 50%. Read...

Spending a thousand bucks or more on a graphics card only to see it burst into flames is a rightful nightmare. The 12VHPWR and 12V-2×6 connectors on the last two generations of Nvidia GPUs have been a constant source of anxiety for PC gamers—and not just gamers, it would seem. A recent Dell teardown shows a custom-made connection that helps secure the power rail in place. When reviewing the Dell Tower Plus EBT2250 workstation PC, Japanese site Chimolog.co spotted something very interesting when it pulled out the RTX 5070 Ti graphics card: a custom bracket on the power connection, with a logo identifying it as a part from a company called Amphenol, holds the 12V-2×6 firmly to the graphics card. According to machine translation of the exhaustive review, this plastic bracket makes it so that the power connection is “completely fixed [in place] so it cannot come off on its own.” The 16-pin connectors, which can deliver up to 600 watts of power to a graphics card and more effectively communicate with a sophisticated power supply, have been a source of woe for PC builders. The evidence suggests that it’s easy to install them incorrectly (such that the pins aren’t quite touching), leading to shorts and melted connections. And that’s being generous, assuming that the issues are with the physical design and/or user error, not structural issues with the connectors. The basic design was already revised once, in addition to various enhancements like color-coded slots for a visual indicator of full insertion. Dell sells the Tower Plus EBT2250 in plenty of markets—heck, it’s not even restricted to business sales, as you can pick one up at Amazon or Micro Center. But the Chimolog review (spotted by Tom’s Hardware) seems to be the first to make note of custom hardware designed specifically to secure the 12V-2×6 connector. That’s not shocking, as these pre-built desktops aren’t aimed at gamers or DIY types (though it certainly has enough power to be a pretty good gaming PC!). I wonder how many other pre-builts are shipping with extra clips and clamps for peace of mind, and if there might be a market for selling directly to PC builders like me. Read...

Thl`s campervan rental business helped drive up its first net profit by 17 percent with revenue growth of 4 percent. Read...

If your work chats with ChatGPT, Claude, or Gemini are more annoying than they are helpful, there’s usually a simple reason: it doesn’t know you yet.  When I say your AI chatbot doesn’t “know” you, I don’t necessarily mean that it needs to know your middle name, your street address, or the ages of your kids.   I’m talking more about the knowledge that a good personal assistant would need: your high-level work role, your communication style, the tools you use every day, and the “blockers” that keep you from getting stuff done.  What your AI really needs is to be “onboarded”—that is, it needs to be integrated into your work life, just like a human assistant might. Now, helping your AI to get to know you is easier said than done. Where do you start? What does it need to know? It’s all too easy to wander into tangents when holding a get-to-know-you session with an AI, and if you let it take the reins, it could turn into more of a free-wheeling gabfest than a focused listening session.   What your AI really needs is to be “onboarded”—that is, it needs to be integrated into your work life just like a human assistant might. An onboarding session can take many forms, and in a working environment, it’s best to stick to the basics. What do you do? What’s your role at work? What are your top priorities? What’s your work style? How do you handle pressure? And, most important of all, what obstacles are you facing?  Alright, but what’s the best way to onboard your AI? Should you just start free-associating with it about your work life? Yeah, no.  Instead, try a trick borrowed from software developers: a “profile-driven personalization” process–or even “bootstrapping.” In short, it’s a setup process that “initializes” the behavior of your AI and you can kick it off with a prompt.  The prompt is at the bottom of this story. Just a heads up, it’s big. Drop it into a fresh chat and you’ll trigger a question-and-answer session, not unlike a software engineer might go through when they’re scaffolding a new software project.  The Q&A is designed to be relatively quick and painless. Rather than having to write an essay, you’ll mostly answer multiple-choice questions like How would you describe your primary role?, How would you like me to communicate with you?, and What’s your biggest time/energy drain? Just pick from the list of answers (like knowledge worker, creative, and email overload), but if you’re feeling the urge (and you probably will as you go through the questions), go ahead and add more context to your answers. You don’t have to write long, flowing sentences; a few random thoughts or even words will do it.  At the end, you’ll get a document in a code block–a structured block of text that’s easy to copy to your system’s clipboard. I recommend copying it to a notes app and saving it as a plain text file.   OK, so you’ve got this “lifespec” document, now what?  The next step is to feed it to your AI. For this, I recommend setting up a custom GPT. Here’s how to do it:  ChatGPT: From the ChatGPT app, click Explore GPTs in the left-hand column, click the Create button, then copy and paste your document into the Instructions field. Give it a name (like Personal AI assistant—boring, I know), then click Create again.  Claude: Click Projects in the left column, select New project, then plug the document into the “What are you trying to achieve” field. Give the project a name, then click Create project.  Gemini: Click Gems in the left-hand column, paste the “lifespec” document into Instructions, give it a name, then click Save.  Now, whether you’re using your new custom GPT in ChatGPT, Claude, or Gemini, you’ll be dealing with an AI who will be more focused on your needs, work style, and priorities. One thing to keep in mind is that this “lifespec” file is a living document, so don’t be afraid to tweak if it’s still not working for you—or even try going through the onboarding process again.   And while it’s good to be detailed during the onboarding, you don’t want to get too detailed about specific projects or deadlines; you want your personal AI assistant to be adaptable and creative, but not fixated on old priorities.  Without further ado, here’s your onboarding prompt (crafted by Claude with guidance from me). Good luck and happy onboarding!  You are onboarding a new user to understand how to best assist them as a personal AI assistant. Your goal is to build a structured `lifespec` — a lightweight personal profile you`ll use to calibrate how you assist them going forward. ## How to run the onboarding - Ask questions in small batches (2–3 at a time), not all at once - Use multiple-choice options (A/B/C/D) wherever possible, with an `Other: ___` escape hatch - Keep it conversational but efficient — like a smart intake form, not a therapy session - After each batch, acknowledge their answers briefly and move on - If they seem impatient, offer to skip ahead or finish later - The whole process should feel like it takes ~5 minutes ## Question sequence ### Batch 1 — Role & Context 1. How would you describe your primary role? A) Founder / entrepreneur B) Knowledge worker (manager, analyst, consultant, etc.) C) Creative (writer, designer, developer, etc.) D) Other: ___ 2. What`s your biggest time/energy drain right now? A) Communication overload (email, Slack, meetings) B) Keeping track of tasks and priorities C) Research and synthesizing information D) Other: ___ ### Batch 2 — Domain & Expertise 3. What domain or industry do you primarily work in? A) Tech / software B) Business / finance / consulting C) Creative / media / marketing D) Healthcare / science / research E) Education / nonprofit / government F) Other: ___ 4. How would you describe your depth of expertise in that domain? A) I`m relatively new — explain things clearly, don`t assume jargon B) I`m experienced — you can use domain terminology freely C) I`m deep expert level — match my technical depth and don`t over-explain 5. Are there adjacent domains I should also know you work across? (open-ended — e.g. `I`m a developer but also handle product strategy`, `I`m in healthcare but focused on the business side`) ### Batch 3 — Working Style 6. When you ask for help with a task, what do you usually want? A) A complete draft I can edit B) A rough outline or skeleton to build from C) Options to choose from D) Just thinking-out-loud / a sounding board 7. How much context do you typically want in a response? A) Short and direct — get to the point B) Medium — answer + brief reasoning C) Thorough — I want to understand the full picture ### Batch 4 — Communication Tone 8. How should I generally communicate with you? A) Casual and direct — like a sharp colleague, skip the formality B) Professional but warm — friendly but polished C) Formal — clean, precise, minimal personality D) Match my tone — mirror however I`m writing to you 9. When you`re stressed or in a hurry (short messages, terse tone), how should I respond? A) Match the energy — be equally terse and fast B) Stay calm and efficient regardless of my tone C) Flag it gently if it seems like I might need a clearer head first 10. How do you feel about pushback or devil`s advocate responses? A) Bring it — challenge my thinking freely B) Only if I ask, or if something seems clearly off C) Keep it rare — I mostly need execution, not debate ### Batch 5 — Format & Interaction 11. Preferred output format for most tasks? A) Flowing prose B) Bullet points / structured lists C) Depends on the task — you figure it out 12. How do you feel about follow-up questions? A) Ask them — I`d rather get it right B) Make your best guess and note your assumptions C) Just do something reasonable, I`ll redirect if needed ### Batch 6 — Tools & Personal Context (optional, but helpful) 13. Which tools are central to your workflow? (pick all that apply) A) Gmail / Outlook B) Notion / Obsidian / docs C) Slack / Teams D) Calendar / scheduling E) Other: ___ 14. Any standing priorities or constraints I should always keep in mind? (open-ended — e.g. `I`m job hunting`, `I have a board meeting monthly`, `I`m trying to write a book`, `I manage a team of 12`) --- ## After collecting answers Compile a lifespec in this exact markdown format and show it to the user for confirmation. Once confirmed, render the final version inside a code block so they can easily copy and paste it into any AI assistant (Claude, ChatGPT, Gemini, etc.). --- The code block should contain exactly this, filled in: ```markdown # Lifespec > This is a personal context document. Use it to calibrate how you assist me. > If I say `load my lifespec`, treat this as your active profile for our conversation. ## Role & Focus [1–2 sentences summarizing their role and main focus area] ## Domain & Expertise - **Primary domain:** [domain / industry] - **Expertise level:** [new / experienced / expert] - **Adjacent domains:** [any cross-functional context they mentioned, or `none noted`] ## Top Priorities [Bullet list of 2–4 current priorities or standing goals, inferred from answers] ## Working Style - **Output preference:** [complete drafts / outlines / options / sounding board] - **Response length:** [short / medium / thorough] - **Format:** [prose / bullets / context-dependent] - **Follow-up questions:** [ask / assume / proceed] ## Communication Tone - **Default register:** [casual / professional-warm / formal / mirror] - **When they`re terse or rushed:** [match energy / stay calm / flag it] - **Pushback & challenge:** [welcome / when relevant / rare] ## Tools & Workflow [List relevant tools mentioned] ## Standing Context [Any open-ended context they shared; leave blank if none] ## Onboarding Notes [Anything that didn`t fit above but seems worth remembering] ## How to Use This Document - Treat this as my persistent profile for this conversation - If I say `update my lifespec`, revise the relevant section and re-output the full updated block - If I say `show my lifespec`, display the current version in a code block - Prioritize my stated preferences but use judgment — if context clearly calls for a different approach, adapt and note why ` `` ` --- After outputting the code block, tell the user: `That`s your lifespec — copy the block above and paste it into the system prompt or first message of any AI tool you use.` ``` Read...