Sep 2, 2026 Below the Harness: Governing a Multi-Model, Multi-Harness World We believe the future is a multi-model, multi-harness world. And we think it needs a new trust model. In 1988, Norm Hardy described a problem that had been quietly breaking systems for years: the confused deputy. A program that takes action using its permissions instead of yours. Today, every AI agent is that deputy. It… Read now
Author: drweb
The foundational language of the web is getting plenty of love lately! While HTML moves a bit more slowly than its buddies, CSS and JavaScript, that tends to be a good thing. I’m not gonna do, like, in here. That’s too basic for you. That shipped in like 2009. You already know it’s a perfect semantic wrapper element “which is intended to be independently distributable or reusable (e.g., in syndication). Examples include: a forum post, a magazine or newspaper article, a blog entry, a product card, a user-submitted comment, an interactive widget or gadget, or any other independent item of…
I was updating my session for the Data API Builder (DAB) recently, mostly to work with the MCP server capabilities. As I was trying to get this working, I realized that the MCP capabilities were newer than my DAB version. I had v1.4Wow, that’s old. The Overview page for the SQL Server MCP site says v1.7 is needed and 2.0+ has bug fixes. I need an update.I had this in my slide deck, but the dotnet tool is used to update the DAB. The command isdotnet tool update –global Microsoft.DataApiBuilderThis showed a download and then in a few minutes it…
Most admins learn usermod -aG in their first week and never read much beyond it. But drop the -a once on a production box, and a developer can lose sudo, Docker, and every other supplementary group they had with a single keystroke. There’s no confirmation prompt and no built-in undo. The usermod command modifies an existing user account. It’s the counterpart to useradd, which creates accounts. Rather than creating a new entry, usermod changes attributes associated with an account that already exists. Everything below was tested on Ubuntu 26.04 with shadow-utils. Where behavior differs on RHEL-based systems, I’ve called it…
Bnei Brak, Israel, August 31st, 2026, CyberNewswireNew capability identifies, attributes and validates API keys, OAuth tokens and other machine credentials stolen from developer and employee endpointsLunar Cyber today announced Token Exposure Monitoring, a new capability designed to identify, attribute and validate Non-Human Identities (NHI) and machine credentials inside infostealer logs, connect them to the affected organization, and determine which exposures require action.The rapid adoption of AI development tools, cloud platforms and automated infrastructure has put a new class of credentials on developer machines: API keys, OAuth tokens, personal access tokens, and other machine identities that provide direct access to valuable…
Patching a single Linux machine is easy. You run its update, reboot if the kernel changed, and you’re done. Patching a whole fleet of these machines? Same process, just repeated more often. Now, in a mixed fleet, machines don’t all get updated the same way. Every Linux distribution comes with its own package manager, the tool that installs and updates software, so you run apt on the Ubuntu and Debian machines; dnf on the Rocky, AlmaLinux, and Oracle machines; and maybe zypper on an old SUSE server nobody wants to touch. But a scheduled job can keep them all current…
Open source supply chain security has been a scanner-and-CVE game for years — pull the package, run the scan, cross-reference known vulnerabilities, hope the metadata is current. That model always had gaps, and AI agents are widening them fast. Agents pull open source components from wherever they can find them, at machine speed, and the after-the-fact scan is not a control that keeps up. What developers and platform teams actually need is a curated catalog that vets components before they enter the environment, not after they cause a problem.Alan Shimel sits down with Leslie Pascual, Field Engineering Manager for AI…
Finding the root cause of slow queries starts by separating work from waits, then testing theories against plans and history.Finding the root cause of slow queries begins with an illustrative Tuesday morning. At 10:12, an order lookup that normally returned in 180 milliseconds took 14.8 seconds. Forty-three sessions piled up within five minutes, and the application team wanted a new index before lunch.CPU stayed modest, logical reads remained familiar, and the cached plan matched yesterday’s fast plan. The request spent almost its whole life on an LCK_M_S wait. A nightly inventory process had left an update transaction open. The visible query was…
The first sign of trouble was a screenshot. We’d just switched on the A/B test via our feature management platform. Within the hour, a senior stakeholder landed in the variant feature flag, opened the app on their own phone, and sent us an image of it. The message, more or less: “Why is there a new tab in my app?”It was a fair question. It was also one I thought we’d answered weeks earlier. Turns out we never really had. That’s the day I learned the difference between telling people and aligning them.The Work We Were Proud OfMy team was…
TL;DR — Key Takeaways Debian rejected both an outright AI ban and a hands-off approach, instead holding contributors responsible for everything they submit. AI-assisted work must be reviewed, tested and corrected by a human, while sensitive data and credentials cannot be shared with third-party AI services without authorization. The policy offers enterprises a practical model: govern risky behavior around AI rather than trying to detect or prohibit the tools themselves. Most organizations writing AI usage policies right now are stuck between two bad options: ban the tools and watch people use them anyway, or say nothing and hope code review…