Browse All Posts (133)

Allison announces a public preview security control for GitHub Enterprise Cloud that requires “proof of presence” (fresh re-authentication or MFA) before users can perform high-impact actions, using Microsoft Entra ID SSO policies to reduce risk from stolen sessions and long-lived tokens.

Burke Holland argues that chat is often the wrong interface for working with LLMs, and shows how GitHub Copilot app “canvases” provide a more task-oriented surface. He walks through examples like a Connect 4 canvas, a winget UI, and a SQLite UI, then connects the idea to automating agent-driven development workflows.

Liam Conroy-Hampton explains how agent extensibility works in Visual Studio Code, including tools and tool sets, permissions and sandboxing, and how MCP (Model Context Protocol) can connect agents to external data sources. He also introduces Agent Plugins for packaging tools and shows how VS Code supports third-party agents like Claude and Codex.

Antonio Morales explains how the GitHub Security Lab’s Taskflow Agent framework can drive an autonomous fuzzing pipeline for C/C++ projects, from harness generation and AFL++ runs to coverage-guided iteration, crash triage, and draft vulnerability reports, with practical guidance on running it safely in disposable environments.

Muzz Imam walks through setting outbound network egress rules for Microsoft AI Foundry hosted agents, using an account-level RAI policy to allow specific FQDN destinations, observe real outbound calls in Audit mode, and then validate blocking behavior under Enforced mode with a controlled probe.

dotnet introduces a practical series on building AI-powered apps with .NET and Microsoft Foundry, highlighting new SDK capabilities, hands-on projects, and the latest Foundry features and services aimed at .NET developers.

Bob Ward explains how Microsoft SQL (SQL Server, Azure SQL, and SQL DB in Fabric) can be used for chat completion and agentic RAG scenarios, including making REST calls to models, building prompts in SQL, and adding security controls like immutable logging, content moderation, and prompt-injection protection.

Alym Rayani rounds up September 2026 updates across Microsoft Security, focusing on securing AI-era workflows: richer investigations in Microsoft Defender with Microsoft Security Copilot, network controls to stop sensitive data reaching shadow AI using Purview and Entra, and new Purview compliance and lifecycle capabilities.

Microsoft Threat Intelligence analyzes Storm-2570, a ransomware affiliate that reuses the same post-compromise tooling and techniques across multiple ransomware ecosystems, and provides concrete mitigation steps plus Microsoft Defender and Microsoft Sentinel hunting queries to detect and disrupt activity before ransomware deployment.

Linda Li announces the general availability of Routines in Microsoft Foundry Agent Service, a managed way to run agents on schedules or in response to external events. The post explains trigger types, how agents can resume long-running work with reminders, and how unattended execution can use either creator identity or an agent identity in Entra ID.

GitHub shares a short clip from the GitHub Podcast discussing the idea that developers may be “mourning” the craft of manually writing software as AI coding tools become more common, and whether traditional coding skills are actually losing value in day-to-day engineering work.

Dan Taylor walks through recent Microsoft Agent Framework updates that make agents more usable in real apps: interactive UIs via AG-UI, longer-term memory via Foundry Agent Service (and Cosmos DB alternatives), CodeAct for multi-step execution, and resilient hosting for recoverable long-running workflows in both Python and .NET.

Fokko at Work demos the GitHub Copilot and VS Code 1.139 updates, focusing on running agent sessions in Dev Containers on remote hosts and several quality-of-life improvements for managing Copilot sessions and chats.

hcamposu explains how to confirm a BizTalk Server 2020 hotfix (KB5091379) has switched the SB-Messaging adapter from SBMP to AMQP before Azure Service Bus stops accepting SBMP on September 30, 2026, using connection checks, ETW tracing, and targeted validation tests.

nitya invites developers to a Foundry Friday AMA focused on improving production AI agents using Azure AI Foundry’s Insights and Agent Optimizer, with a live demo and Q&A on turning trace patterns into better instructions, tools, skills, and model choices.

Karl-WE curates a practical hub of tooling and reference links for teams evaluating or operating Azure Local, covering community GitHub projects, Microsoft-supported utilities for sizing and troubleshooting, and OEM documentation from Dell and HPE for certified hardware deployments.

Jan-Kalis announces the general availability of Azure Container Apps Sandboxes and explains how they provide per-task, hardware-isolated microVMs for running untrusted code and agent workloads. The post covers network egress controls, VNet/private endpoint connectivity, snapshots and lifecycle policies, persistent volumes, and telemetry export options.

Microsoft Developer announces a Copilot Studio skilling session on October 8, 2026, focused on what to learn first, how to get started building with Copilot Studio, and how Agent Academy supports hands-on AI builder skills.

Microsoft Developer shows how to use the Microsoft Foundry model catalog to choose an AI model based on measured results instead of intuition, comparing deployments with the same evaluation while tracking quality, latency, token usage, and cost.

Allison announces generally available updates to GitHub Copilot code review, adding a dedicated personal settings page and an enterprise-wide default review effort setting for organization-owned repositories.

Allison announces that Node.js 20 has been removed from GitHub Actions runners and that JavaScript actions now run on Node.js 24. The post explains what action maintainers and workflow users need to update, and calls out runner compatibility breaks for older macOS and ARM32 self-hosted environments.

The NuGet Team explains how to prepare for Microsoft’s NuGet author-signing certificate rollover on September 23, 2026, including who is impacted and the exact commands and configuration changes needed to keep trusted-signer policies and package verification working.

Alberto Gimeno explains how the GitHub Copilot app was reworked to keep pull request reviews responsive even with million-line diffs and hundreds of inline comments, focusing on virtualization limits, dynamic height measurement, scroll anchoring, and the instrumentation loop used to harden performance.

Simon Frost explains how the Microsoft Discovery app and Premonition Insights on Azure can speed up exploratory analysis of wastewater metagenomics, while keeping human scientific judgment in control. The post walks through an end-to-end workflow: literature grounding, taxonomic filtering, weather-data enrichment, phylogenetic verification, and interactive dashboards.

abbyhamilton highlights several Catalyst series customer stories showing how organizations use Microsoft Azure AI infrastructure alongside NVIDIA to run high-performance AI workloads, including real-time weather forecasting, multimodal health screening, large-scale biological databases, photorealistic digital twins, and rare-disease patient identification.

Mark Russinovich discusses what resilience means at Azure’s scale, using a 2014 near-outage as a starting point to explain how Azure tests failure, defines service health, and responds to incidents. He also covers how AI and AI agents change reliability assumptions, and where Azure’s architecture frameworks fit.

Mark Russinovich, Adam Bogobowicz and Molina Sharma explain why resilience on Azure can’t be proven by an architecture diagram alone, and why teams need continuous validation as workloads change—especially when AI models and other probabilistic dependencies become part of the critical path.

Rob Lefferts announces Integrated Security Operations Center (ISOC) in Microsoft Defender, positioning it as a unified foundation for “agentic” security operations where protection and SOC workflows run as one system. The post explains the stack layers (signals, context, actuators) and how an integrated protection loop can speed up detection, investigation, and response.

simonjj announces the general availability of Azure Container Apps Express, a streamlined way to deploy a container image as a production-ready app on Azure with instant provisioning, scale-to-zero, and sub-second startup optimizations, plus broad regional availability and a sandboxed compute foundation.

Meagan demonstrates how GitHub Copilot can be used across Slack, Microsoft Teams, GitHub.com, the GitHub mobile app, and VS Code to keep coding work moving in the same conversation where teams collaborate.

Allison announces a public preview feature in the GitHub Copilot app that lets teams sandbox local sessions per project to reduce the blast radius of unintended commands. It covers filesystem allow/deny lists, outbound and local network controls, and limiting access to Git and GitHub CLI credentials.

Mike Hulme explains what changes when software shifts from request/response apps to continuously running AI agents, and why “agent-first” platforms need governance, identity, tracing, and runtime guardrails. The post connects Microsoft Foundry governance with Azure Container Apps Sandboxes for isolated, per-task execution at scale.

Visual Studio Code highlights the Debug Visualizer extension for VS Code, showing how it can visualize data structures while debugging so you can watch algorithms evolve as your program runs.

Simon Frost explains how he used the Microsoft Discovery app to speed up exploratory analysis of a large wastewater metagenomics dataset, while keeping human scientific judgment in the loop. The post walks through agent-assisted data wrangling, API enrichment, phylogenetics, and visualization to produce auditable, reproducible findings.

Paull Young shares findings from a GitHub and Yale survey of 1,039 GitHub users on sustainable software, highlighting demand for practical tools, credible measurement, and workflow-friendly ways to reduce wasted compute—especially in repositories and CI/CD.

John Edward covers Microsoft and AWS’s Azure Multicloud Interconnect for AWS announcement, focusing on what it changes for private, high-bandwidth connectivity between the two clouds and why it matters for teams running cross-cloud, data-heavy workloads.

Thomas Maurer argues that Azure Local should be evaluated as the infrastructure foundation of Microsoft Sovereign Private Cloud, not just as a hypervisor replacement. He breaks down connected vs disconnected operations, Azure Arc control-plane options, and the platform capabilities (AKS, data services, Foundry Local, GitHub Enterprise Local) that shape sovereignty and resilience.

The Visual Studio Code Team shares an update for the VS Code 1.140 Insiders build, highlighting a change to remote agent chat that lets you start a chat session on an eligible remote agent host without first selecting a folder in the desktop Agents Window.

Lee Stott walks through Fibey Field Ops, a synthetic field-service assistant built with Microsoft Foundry Hosted Agents, MCP, and Azure Container Apps, focusing on the practical engineering work around tool contracts, identity boundaries, deployment sequencing, and debugging/observability rather than just model choice.

Allison announces OpenTelemetry support in the GitHub Copilot app, letting enterprises export agent activity telemetry to compatible observability tools so teams can trace agent sessions, investigate unexpected behavior, and manage monitoring settings centrally.