Browse All Posts (101)

Allison announces a public preview update for GitHub Projects and GitHub Issues: multi-select fields, allowing a single field to store multiple values and enabling filtering, grouping, and searching across project views.

Peter Ibekwe announces Agent Framework Declarative Workflows 1.0 for both Python and .NET, showing how to define multi-agent orchestration in YAML (branching, state, handoffs, and human approvals) and then load and run it as a standard runtime workflow.

GitHub hosts a “Rubber Duck Thursdays” coworking session where viewers can hang out with the team, ask questions, and hear what’s new in the GitHub ecosystem.

HARMEETGILL announces the general availability of OneLake resource instance rules in Microsoft Fabric, explaining how workspace admins can allow secure service-to-service access from specific Azure resource instances (like Databricks or Azure SQL) without relying on broad IP allowlists or private networking everywhere.

John Savill gives a quick, practical introduction to Windows Sandbox, showing how to install it, what the underlying architecture looks like, and how to customize a sandbox session for safely running apps or processes in an isolated Windows environment.

Microsoft Developer introduces SQL MCP Server 2.0 and explains how it helps agentic apps work with production SQL data more safely, including new OBO authentication, MCP tool improvements, and security guardrails, with demos covering configuration and controlling access across Graph, REST, and MCP.

Anna Hoffman introduces Change Event Streaming (CES) in public preview across Microsoft SQL, showing where it fits for modern, event-driven application scenarios and how it differs from approaches like CDC and mirroring in SQL Server 2025, Azure SQL, and Fabric SQL.

Carlin Cherry explains why Dependabot now applies a default three-day cooldown before opening non-security version update pull requests, aiming to reduce exposure to fast-moving supply chain attacks where malicious package releases are discovered and removed within hours.

GitHub shows how to tailor the GitHub Copilot for Jira integration so teams can reach draft pull requests faster, using ticket-level and space-level instructions to set defaults like target repo and branch naming, and to route specific work (such as API refactors) to custom agents.

Microsoft Threat Intelligence and the Microsoft Defender Security Research Team summarize Q2 2026 email and Teams threat trends, including the post-disruption decline of Tycoon2FA-linked phishing, shifts in QR code and CAPTCHA-gated lures, and notable BEC and malware-delivery campaigns, plus concrete Defender configuration and response guidance.

James demonstrates how the GitHub MCP Server connects GitHub Copilot to GitHub workflows inside VS Code, enabling natural-language issue management, pull request creation, and code review without switching tools.

Visual Studio Code shares a short clip highlighting a GitHub Copilot-related “did you know?” tip in VS Code, framed as a quick GitHub “hacking” moment for developers.

GitHub experiments with building a personal assistant using the GitHub CLI, focusing on how to wire up workflows and try out GitHub features from the command line.

junjieli introduces the Foundry Agent Canvas (public preview), a GitHub Copilot App extension that lets developers discover resources, scaffold and configure a Foundry hosted agent, test it locally with an embedded Agent Inspector, and deploy it to Foundry Agent Service using azd-driven workflows.

anandgmenon walks through running Azure Logic Apps Hybrid on a self-managed RKE2 Kubernetes cluster, including how to provide an ingress IP with MetalLB, connect the cluster to Azure Arc, install the Container Apps extension, and fix the RKE2-specific DNS and inotify issues that can break deployments.

Authorised Territory demonstrates how to add sampling to a .NET Model Context Protocol (MCP) server so an MCP tool can call back into a client’s locally running LLM (via Ollama) using a chat client.

brauerblogs shares a reminder to register for Microsoft’s “Path to Production for Agents” webinar series (July 27–28), focused on taking AI agent solutions from experimentation to secure, scalable production with guidance on governance, platform design, AgentOps, and multi-agent architecture patterns.

saniya0307 explains how Azure Local adds iSCSI SAN integration alongside Fibre Channel, enabling hybrid (S2D + SAN) and SAN-only disaggregated deployments so organizations can reuse existing storage arrays, scale compute and storage independently, and modernize virtualization infrastructure with Azure-consistent management.

avinandac introduces the generally available Lakehouse Query Explorer in Microsoft Fabric, an in-context Spark SQL editor inside the Lakehouse explorer for quick data exploration, validation, and iteration, with guidance on when to switch to notebooks or the SQL analytics endpoint for more advanced workflows.

Andrea Griffiths breaks down what GitHub Copilot subscriptions pay for beyond raw model API calls, including the developer workflow surfaces Copilot connects, how GitHub AI Credits are metered, and when direct API access (or BYOK) makes more sense for teams building their own agent systems.

The Visual Studio Code Team shares the ongoing release notes for VS Code 1.131 (Insiders), including links to the commit log and closed issues so developers can track what’s landing in the build as features are added.

Kalaivanan explains how to use Azure API Management (APIM) as a control plane for Model Context Protocol (MCP) servers, focusing on enterprise-ready authentication, access control, observability, and governance. The post lays out practical patterns for putting APIM in front of MCP endpoints and using Entra ID, OAuth flows, and API Center for discovery.

Allison announces a new GitHub Copilot metrics impact dashboard for enterprise admins and org owners, turning Copilot usage metrics API cohort data into a visual view of adoption depth, developer impact, and recommended next steps to move users into deeper adoption phases.

Catherine Cassell outlines changes to GitHub’s bug bounty program, including a permanent invite-only VIP track, new static payout tables for both VIP and public researchers, and a HackerOne signal requirement aimed at reducing low-effort and AI-generated submissions.

Aarti Borkar explains how Microsoft and AXA XL are partnering to make Microsoft Defender Experts Cybersecurity Incident Response available to AXA XL cyber insurance policyholders, with an emphasis on coordinating technical containment and recovery work alongside legal, executive, and insurance decision-making during high-stakes incidents.

Allison announces a GitHub Enterprise Server security change that will reject command-line support bundle uploads from unpatched GHES versions starting August 18, 2026, and lists the minimum patch releases required to avoid disruption.

Linda Li and Maria Naggaga explain how Azure AI Foundry Toolboxes let agents call MCP tools using the signed-in user’s identity, without embedding OAuth/token plumbing in agent code. The post covers per-user token isolation, consent handling, and governance features like RAI guardrails and gateway fronting.

GitHub demonstrates how to kick off GitHub Copilot work directly from Jira to produce a draft pull request, including three ways to trigger Copilot and what the review-ready output looks like in GitHub.

Wes Steyn announces the release of the Microsoft Agent Framework Harness, a batteries-included runtime that turns a chat model into a production-ready agent with planning, tool-calling, memory, approvals, web search, and OpenTelemetry built in for both .NET and Python.

Ricardo Duncan describes how CDK Global migrated a business-critical automotive CRM platform to Azure SQL Managed Instance, including the architecture choices, migration approach, and operational practices used to move more than 1,000 databases with minimal downtime.

Brady Gaster joins GitHub’s Open Source Friday to introduce Squad, an open source project aimed at building teams of AI agents that can collaborate on work for any kind of project.

Gloridel Morales announces July 2026 patch releases for Azure DevOps Server, including download links for Patch 6 and Azure DevOps Server 2022.2 Patch 11, plus a quick command to verify whether the patch is installed on your server.

GitHub shows how the GitHub Copilot app can run multiple AI coding agent sessions in parallel, so different parts of a codebase can be worked on at the same time. The example highlights splitting work across agents for bug fixes, test writing, and feature development without clobbering each other’s changes.

Visual Studio Code shows how to use Oh My Posh to customize and improve the terminal prompt inside VS Code, with a link to the Oh My Posh project for setup and themes.

richpaw describes a reference architecture for running Microsoft Discovery on a Windows VM in Azure and connecting it to an Azure CycleCloud HPC cluster via Azure NetApp Files (NFS) and SSH, so agentic workflows can submit Slurm jobs, read/write POSIX files, and stay inside a private network boundary.

richpaw explains how Microsoft Discovery combines agentic AI with Azure HPC to speed up semiconductor EDA workflows, focusing on closed-loop orchestration across tools, data, and compute so teams can triage regressions, explore physical design tradeoffs, and converge on signoff faster.

Ron Frenkel explains how the Azure Copilot Observability Agent (now generally available in Azure Monitor) helps teams investigate Azure AI Foundry and GenAI agent issues using Application Insights telemetry, including failures, latency, tool-call errors, token spikes, and dependency bottlenecks with evidence-backed root cause analysis.

Jacklyn Carroll introduces canvases in the GitHub Copilot app: shared, interactive workspaces that let you collaborate with an agent beyond a chat thread. The article shows example prompts and outcomes for building canvas extensions that support issue triage, codebase visualization, session/worktree cleanup, prompt coaching, and cross-tool knowledge discovery.

GitHub gives a beginner-friendly tour of the GitHub Copilot app, positioning it as an agent-native desktop workspace for GitHub tasks. It shows how to start an agent session, hand off a real coding task, watch Copilot explore a codebase, run tests, and iteratively review and merge changes.

Allison announces that Gemini 3.6 Flash is rolling out as a selectable model in GitHub Copilot, highlighting its focus on coding and longer-horizon agentic tasks, plus where it’s available and what admins need to enable for Business and Enterprise plans.