Microsoft
Developer Blogs
Get the latest information, insights, and news from Microsoft.
Highlights
Building Agents for Teams: Turning conversations into outcomes
How to test agent skills without hitting real APIs
Latest posts
Azure SDK Release (July 2026)
Azure SDK releases every month. In this post, you'll find this month's highlights and release notes.
Tell your model when to think harder
Not every question deserves the same amount of thought. Renaming a variable isn't the same as debugging a memory leak, and they don't need the same level of thinking. So why should your model treat them the same way every time? Starting in Visual Studio 18.9 Insiders 2, it doesn't have to. Supported models now come with a thinking effort control, so you can dial up the reasoning when a problem is genuinely hard and dial it back down when it isn't. It's a small knob with a big payoff: better-matched answers, and more control over how many credits you spend getting them. A dial, not a switch Thinking effort is...
Making an agile version of a Windows Runtime delegate in C++/WinRT, part 8
Exceptions hiding in lambda captures.
Announcing v2.0 of the official MCP C# SDK
MCP C# SDK v2.0 implements the 2026-07-28 specification with a stateless-first protocol, standardized HTTP headers, and Multi Round-Trip Requests for interactive tools, all while staying backward compatible.
Package Integrity and Enforcement
A recent conversation started (as many interesting conversations do) with a simple question: "Why is MSIX secure?" MSIX includes multiple layers of protection designed to defend against both accidental damage and malicious tampering. But as with all good security solutions, it's not the individual elements so much as their interlocking design that makes the whole stronger than the sum of its parts. MSIX was designed around several core facets, integrated in sometimes novel ways. A packaged process can be trusted to a higher degree than most applications because Windows maintains a verifiable chain of trust fr...
Visual Studio July Update — Meet the New Agent, Powered by the GitHub Copilot SDK
There's a new Agent (Preview) in Copilot Chat, built on the same GitHub Copilot SDK that powers the GitHub Copilot CLI, that gets more done with less back-and-forth.
Find any commit in seconds: Commit Search comes to Azure DevOps
Great code has a long memory. The fix you shipped last quarter, the refactor that touched a dozen files, the one-line change that explained why something works the way it does — it's all there in your commit history. The hard part has always been finding it. Today, we're excited to announce that Commit Search is entering general availability (GA) in Azure DevOps. This is a capability our team has been building and refining for quite some time — through design, a private preview with early customers, and a careful, staged rollout — and we couldn't be happier to bring it to everyone. Why we built Commit Search C...
Making an agile version of a Windows Runtime delegate in C++/WinRT, part 7
Further explorations into exception safety.
Refactor your database with SQL projects in VS Code
SQL database projects are a fundamental tool for keeping your database in source control, tracking changes and collaborating on quickly evolving database needs. Not to be forgotten, their superpower is being able to dynamically generate deployment scripts that match the declared state and the exact environment you're deploying to. SQL project development becomes more challenging when a database needs serious changes, where the requirements of the system have evolved to the point of having to rename tables and move objects between schemas. In plain T-SQL, we can rename a table or column without having to co...
Discover Agent Skills from MCP servers in .NET
Your agents can now discover and load Agent Skills directly from a Model Context Protocol (MCP) server. Instead of shipping every skill inside your application or copying skill folders into each deployment, you point an agent at an MCP server and it pulls the skills it needs on demand. A central team can publish skills once, and every agent across your organization picks them up without a redeploy. This is available today in .NET through the package. For makers, this removes a distribution problem: you author a skill in one place and serve it to many agents. For enterprise leaders, it means domain expertise –...
Query Performance Tuning in Azure DocumentDB: A Practical Guide
Writing a query that works is easy. Writing a query that scales is a different story. As collections grow to millions of documents, queries that returned results in milliseconds during development start taking seconds or timing out entirely. The root cause is almost always the same: the database is doing far more work than it needs to. In this post, we’ll take a single, realistic query and walk through the full optimization process: running explain() to understand what’s happening under the hood, building indexes step by step, and applying the ESR rule to design a compound index that eliminates wasted scans and...
Analyze MSBuild Binary Logs with Copilot in VS Code
Meet the MSBuild Binlog Analyzer for VS Code - a Copilot-powered way to read MSBuild binary logs, explain and fix build failures with one click, compare builds, and catch regressions, backed by the Microsoft.AITools.BinlogMcp MCP server.
A new way to bring native Windows APIs to JavaScript – introducing dynamic API projections for Node.js
Electron and Node.js make it straightforward to build Windows desktop apps in JavaScript. Calling Windows Runtime APIs has been less straightforward: features such as on-device AI often required a C++ or C# bridge, manual translation of WinRT types and asynchronous behavior, and wrapper code for every exposed API. That meant another language and toolchain to build, test, and keep compatible with Electron. We're making a dynamic Windows Runtime API (WinRT) projection for Node.js available in public preview. It lets an Electron app or plain Node.js process call supported Windows Runtime APIs directly fro...
Making an agile version of a Windows Runtime delegate in C++/WinRT, part 6
Exception-safety, the invisible bug.
Java OpenJDK July 2026 Patch & Security Update
Hello Java customers! We are happy to announce the latest July 2026 patch & security update release for the Microsoft Build of OpenJDK. Check our release notes page for details on fixes and enhancements or download and install the binaries today. The source code of our builds are now available on GitHub for further inspection: Microsoft Build of OpenJDK specific updates OpenJDK 25 OpenJDK 21 OpenJDK 17 OpenJDK 11 Summary of Upstream Updates OpenJDK Vulnerability Advisory July 2026 OpenJDK 25 OpenJDK 21 ...
Making an agile version of a Windows Runtime delegate in C++/WinRT, part 5
Making sure to use the non-agile delegate non-agile-ly.
Native Agent Memory for Microsoft Agent Framework, Powered by Azure Cosmos DB
Recently we introduced the Agent Memory Toolkit and the Agentic Retrieval Toolkit for Azure Cosmos DB. The Agent Memory Toolkit gives your agents durable, Cosmos-backed memory: it stores raw conversation turns and then distills them into higher-value derived memories (thread summaries, extracted facts, and cross-thread user profiles), all searchable with vector, full-text, and hybrid search in the one database you already use. Today we're taking the next step. With the latest release of Microsoft Agent Framework, you can now drop that memory into an agent with a single object: the , shipped in the new package f...
The Microsoft 365 Copilot Agent’s Playbook: A Practical Livestream Series for Building Better Agents
Building on Microsoft 365 Copilot? Here's your playbook. Declarative agents are quickly becoming one of the most exciting ways to extend Microsoft 365 Copilot and bring organizational knowledge, workflows, and tools directly into the flow of work. But as agent capabilities grow, so does the need for practical guidance: How do you build agents that are useful, grounded, extensible, and measurable? That’s why we’re launching The Microsoft 365 Copilot Agent’s Playbook, a four-part livestream series designed to help developers, makers, and technical teams understand how modern Microsoft 365 Copilot de...
Move Agent Orchestration/Workflows out of Code with Agent Framework Declarative Workflows 1.0
Most multi-agent apps wire every flow in application code: the sequence of steps, branching, and handoffs between agents all live inside the program, making the orchestration harder to review, version, and change. Declarative workflows make that orchestration explicit. In YAML, you define how agents coordinate, how state changes, where execution branches, and when people step in. Agent Framework loads the definition into a standard workflow you can run, stream, and compose with code. Today, declarative workflows reach 1.0 across both Agent Framework SDKs. Python's package is now 1.0.0, joining th...
Build locally, ship to Azure: meet Azure SQL Developer
Big news: Azure SQL Developer is here, in private preview. It's the Azure SQL Database engine, on your laptop, in a container. Build against the exact engine you run in the cloud. Ship the same code to Azure. Change one line, the connection string, and you're in production. Free for local dev and CI. No subscription. No credit card. No catch. Run it yourself, or hand it to an AI agent and watch it go. The inner-loop problem If you build apps on a cloud database, you know the friction. To develop and test locally, you either point your app at a shared cloud instance, with slow round trips, noisy neighbors, and ...
Making an agile version of a Windows Runtime delegate in C++/WinRT, part 4
Optimizing the context check.
Smarter Spring Development in Eclipse with GitHub Copilot
How the Spring Tools MCP Server and GitHub Copilot together make Spring Development more efficient. Introduction If you write Spring Boot applications in Eclipse, you already know that Spring Tools for Eclipse brings best-in-class tooling right into your IDE, including quick navigation and visualization of core Spring elements in your projects, live information from running Spring Boot applications, Spring specific auto completion, validations and quick fixes, request mapping overviews, and much more. In addition to that, if you have the GitHub Copilot plugin for Eclipse installed, you have a powerful AI codi...
Pure Virtual C++ 2026 Is a Wrap
Pure Virtual C++ 2026 is a wrap. Every featured and on-demand session is now available on YouTube — catch up on C++/Rust interop, AI-driven tooling, build performance, and more.
Windows App Development CLI v0.5.0 – expanded UI automation, JS/TS bindings, and more
WinApp CLI v0.5.0 greatly expands the UI automation toolkit (recording, touch, pen, and keyboard injection), lets you call Windows Runtime APIs straight from JavaScript and TypeScript with zero native addons, and makes WinUI crashes far easier to diagnose. Get the update by running winget install Microsoft.WinAppCLI or check the repo for other install options. Here's what the new version has to offer in depth: 🕹️ An expanded UI automation toolkit The winapp ui command family grew from inspection into a complete automation toolkit. On top of the existing inspect/click/screenshot verbs, these relea...
Building Agents that Act on Your Behalf with Toolboxes in Foundry
How Toolboxes in Foundry simplify user delegation At some point, many agents move from answering questions to taking action. And when it does, the question becomes: whose identity is it acting with? Imagine you are building an internal employee agent. It needs to call a private, Entra-protected MCP server for orders, and it also needs to use Microsoft’s managed Work IQ MCP server to reason over the employee’s Microsoft 365 context. In both cases, the agent can’t run as a managed identity or service account. It has to act as the real signed-in user — with that user’s permissions, access boundaries, and data pr...
Making an agile version of a Windows Runtime delegate in C++/WinRT, part 3
The object that actively refuses to be marshaled.
The Microsoft Agent Framework Harness is now released
Your agents can now be built on a stable, batteries-included harness - the loop, planning, memory, context management, approvals, and telemetry that turn a model into an agent that actually does things - in both Python and .NET. What is an agent harness? An agent harness is the scaffolding that turns a language model into an agent. A model on its own can only generate text. To have it call tools, work through multi-step tasks, remember what it has done, and keep going until the job is finished, you need a runtime wrapped around the model - and that runtime is the harness. Agent Framework ships a ready-made o...
Pure Virtual C++ 2026 Is Now Live!
Pure Virtual C++ 2026 is streaming now — watch five featured sessions live with Q&A, plus a full on-demand playlist, all free.
Introducing the ETW MCP: AI-assisted ETL trace analysis, headless and in your terminal
If you’ve ever opened an ETL trace with hundreds of data tables and wondered “where do I even start?” This post is for you. We’re releasing an early preview of the ETW MCP, a Model Context Protocol server that lets GitHub Copilot, or any MCP-aware AI assistant, read, query, and reason over Event Tracing for Windows (ETW) traces the same way an experienced engineer would. No UI required. This is a companion to the WPA MCP (coming soon), which brings the same AI-assisted experience inside Windows Performance Analyzer. The ETW MCP server is the headless sibling, same data layer, no WPA needed, works anywhere you ...
July Patches for Azure DevOps Server
We are releasing new patches for our self‑hosted product, Azure DevOps Server. We strongly recommend that all customers stay up to date with the latest, most secure version of Azure DevOps Server. The most recent release, Azure DevOps Server, is available on the download page. The following versions have been patched. For more details on these updates, see the release notes: ⬇️Azure DevOps Server Patch Download ✅Verifying Installation To verify that the patch is installed, run the following command on the Azure DevOps Server machine using the patch installer you downloaded: Replace with the name of the...
Display Name is for Humans
Package identity exists for one primary purpose: This is a considerable improvement over the historical tendency to assign UUIDs and GUIDs to nearly everything - including things that humans were expected to discuss in meetings. Identity Is for Systems A package's Name, Package Full Name, and other identity-derived values are intended for developer and system use. These identifiers prioritize correctness, uniqueness, and programmatic consumption - not clarity or friendliness from an end-user's perspective. Identity must remain stable across servicing operations and dependency resolution. If it changes, the...
Making an agile version of a Windows Runtime delegate in C++/WinRT, part 2
Short-circuiting the easiest case.
How to test agent experience changes without shipping them
Most changes you think will improve AI agent behavior won't. We tested a dozen hypotheses on a real project upgrade scenario and the majority failed. Learn how to emulate documentation, API, and MCP server changes locally so you can validate what works before shipping anything to production.
C++ Dependencies Without the Headache: vcpkg + Copilot CLI
At Pure Virtual C++ 2026, we build a C++ console app from an empty folder using CMake, MSVC, and GitHub Copilot CLI. This walkthrough shows a practical prompt sequence for dependency selection, reproducible builds, and controlled updates.
Pure Virtual C++ 2026 Is Tomorrow and On-Demand Sessions Are Now Available
The on-demand sessions for Pure Virtual C++ 2026 are available now on YouTube. Watch seven talks on SPGO, vcpkg + Copilot CLI, CMake Tools, PackageReference, MSVC upgrades, and C++23/26 status — then join us live tomorrow.
Announcing a New Visual Manifest Editor in the WinApp VS Code Extension
The WinApp VS Code extension exists to make Windows app development feel at home in VS Code. It brings the Windows App Development CLI right into the editor, so you can initialize, run, debug, package, and sign Windows apps built with .NET, WPF, WinUI, C++, Electron, Rust, Tauri, or Flutter, all without switching tools. Get it now: Install the WinApp extension from the VS Code Marketplace. We're excited to announce the release of v0.2. This release includes a new editor that supports editing a manifest file without touching raw XML and support for running WinApp commands in multi-app workspaces. A Visu...
T-SQL Hygiene: Introducing the Covering Index
Often in applications, we write database queries. And often, the same query is executed again and again. To make queries against large tables faster, we can add an index. This is a general improvement for anyone accessing the table. However, there is a specially designed index called a covering index that can make queries against large tables particularly faster for one specific query. We call it covering because it covers all the projections and predicates in that query. Which query? That is up to you, the developer, to identify. Look for queries that are either 1) very important to run fast or 2) extreme...
AI agents, meet the Azure Cosmos DB vNext emulator
If you use the Azure Cosmos DB vNext emulator, you probably know the local development loop: start the emulator, connect to it, create some resources, load test data, run queries, and inspect the results. Each step is straightforward, but together they add setup work before you can test the application you are actually building. How agents work with the emulator The emulator includes the Azure Cosmos DB Shell, an open-source CLI for working with databases, containers, and items. It runs inside the emulator container and handles the local endpoint and well-known key, giving developers a direct, scr...
Making an agile version of a Windows Runtime delegate in C++/WinRT, part 1
The easy case is easy.
Faster C++ iterative builds with GitHub Copilot
Slow builds are a consistent theme of feedback from C++ developers. We built GitHub Copilot build performance for Windows so you can leverage Copilot to optimize your project’s build times. This workflow will find optimizations that bring your build times down. At first, we only measured the impact of build optimizations on full clean builds. While that number is valuable, we know that you are not rebuilding from scratch on most days. You edit a file and build, then repeat the process as part of your iterative loop. Working with customers after the public preview, we noticed that every now and then, the age...
Why has the display control panel pointer truncation bug gone unfixed for so long?
It's fixed, but the fix isn't getting there.
How to test agent skills without hitting real APIs
Your agent skill calls an API. The moment you start evaluating it, every run either costs money or mutates production data. Learn how to mock APIs transparently so you can run evals without changing your skill or hitting real endpoints.
Teaching a Vision Model to See Like a Human Annotator—and Catching It When It Lies
A multimodal LLM enrichment pipeline that extracts structured metadata from visual assets, constrains output to predefined values to minimize hallucinations, and uses a ground truth evaluation template to measure quality—all as a plug-and-play module.
Pure Virtual C++ 2026 [Meet the Speakers, Part 3]: Modernizing C++
Meet the speakers behind Pure Virtual C++ 2026. In the final part, Victor Ciura maps the real challenges and promising directions for C++/Rust interop.
Build Once, Run Everywhere: Unified Manifest for Office Add-Ins now Generally Available
Unified manifest support for Word, Excel, PowerPoint, and Outlook makes it possible for developers to build a single app that works across Microsoft 365 apps, creating a more consistent experience for users and a simpler deployment model for IT admins.
T-SQL Hygiene: What’s the Big Deal with SET NOCOUNT ON?
Learn why SET NOCOUNT ON improves stored procedure efficiency, reduces unnecessary messages, and matters to .NET developers.
Announcing .NET Modernization for Beginners
A free, open-source, hands-on course that walks you through modernizing a real legacy ASP.NET application all the way to .NET 10 using the GitHub Copilot modernization agent, step by step.
How to build long-running MCP tools on Azure Functions
Learn how to build long-running MCP tools on Azure Functions using Durable Functions. This post explains why synchronous tool calls break down for long-running work, introduces the MCP Tasks extension, and walks through a sample pattern that works with existing request/response clients today.
Visual Studio Administrator? Join our Private Marketplace Preview!
Organizations are increasingly looking for greater control over extensions within development environments. Driven by security, compliance, and internal governance requirements, teams want more visibility into how developers discover and acquire extensions. To address these needs, we're excited to begin previewing Private Marketplace support in Visual Studio. Private Marketplace for Visual Studio For organizations familiar with Private Marketplace in VS Code, Private Marketplace in Visual Studio provides a similar curated extension acquisition experience while preserving the familiar Visual Studio workfl...
Speculating on how the buggy control panel extension truncated a value that it had right in front of it
Inferring the code's history.
MSVC Build Tools Preview updates – July 2026
Try the latest updates to MSVC Build Tools Preview
MSSQL Extension for VS Code: Shortcuts Configuration, Enhanced Results Grid Experience, and More
The MSSQL extension for VS Code v1.44 continues to improve the SQL development experience with features that help you work faster and customize your workflow. In this release, we're introducing the Public Preview of Shortcuts Configuration and an Enhanced Results Grid experience. Together, these updates make it easier to personalize how you work with SQL and improve the performance and usability of query results directly inside Visual Studio Code. What's new in MSSQL extension for VS Code v1.44 Here's a summary of the key features in this release: Shortcuts Configuration (Preview) The new Shor...
Agent Skills for Python Is Now Released
Your Python agents can now pick up reusable bundles of domain expertise (instructions, reference material, and scripts that load only when a task calls for them) through a stable, production-ready API. Agent Skills for Python in Microsoft Agent Framework is stable and shipping: the core skills API has no experimental gate, so you can rely on it in production without the churn of a preview surface. Teams can author skills, release them on their own schedule, and drop them into any agent, backed by the governance controls enterprises expect before agents reach production. If you've been following our earlier...
Pick, manage, and get the most from your models
You open the model picker, scroll past a dozen options, and pause. How are these models different? Which one should you actually use? And once you're a few hundred messages deep, how much capacity is even left before things start dropping off? We've all been there. These are the kinds of questions Visual Studio now makes a little easier to answer. Here's a closer look at how you can see, compare, and get more from the models you already have. A model picker that works the way you do The model picker presents a long list of models you’ve needed to scroll through every single time. Starting in 18.9 Insider...
The case of the invalid function pointer when shutting down the display control panel
Watching the bits disappear.
Building AX evals that actually work
This is the eighth and final article in a series about Agent Experience (AX): the practice of making AI coding agents work correctly with your technology. The series covers what you can and can't control in the agent stack, how to measure whether your extensions are helping or hurting, and how to iterate toward better outcomes. You've read seven articles about what to measure, why benchmarks don't transfer, and what hidden variables can do to your results. Now you actually have to build the thing. Most teams build an eval, run it, get scores, and feel good about the numbers. The trouble is that these evals produ...
Rethinking C++ Performance: Faster Code Navigation and GitHub Copilot Tools with Whole Codebase Indexing
In large C++ codebases, your code understanding and navigation depend on quickly determining how symbols, declarations, definitions, and references are connected across your project. In Visual Studio Insiders 18.9, the new whole codebase indexing (WCI) enhances the existing browse database via a deeper, more comprehensive indexing approach. This preview feature allows Visual Studio to access richer symbol information more efficiently in your C++ project, with faster lookups for core IntelliSense scenarios like Find All References and semantic file colorization. It also enables new experiences like CodeLens ref...
.NET and .NET Framework July 2026 servicing releases updates
A recap of the latest servicing updates for .NET and .NET Framework for July 2026.
CoreCLR Progress and the Mono Timeline for .NET MAUI
CoreCLR is now the only runtime for .NET MAUI mobile apps in .NET 11 Preview 6. Here is where we are, and why we need you to test today.
.NET 11 Preview 6 is now available!
Find out about the new features in .NET 11 Preview 6 across runtime, SDK, libraries, ASP.NET Core, .NET MAUI, C#, Entity Framework Core, F#, and container images.
Built-in Agent Skills Bring .NET and Azure Expertise into Visual Studio
Visual Studio now includes built-in Agent Skills, created by experts from the .NET and Azure teams, to help you better customize your agentic workflow and complete development tasks more efficiently, starting with the 18.8 Release. Agent Skills are reusable capabilities that enable your agent to perform structured tasks more reliably (to learn more about what are Agent Skills, see this previous post). We’ve heard that getting started with skills can feel unclear, especially when deciding which ones to use and how to apply them. To simplify this experience, we’ve introduced a set of built-in skills for common ....
Package Status
Every package has a set of status flags describing its current health and availability. These can be queried via the Package.Status property. A package may report more than one status value at a time if multiple conditions apply. Package Status Groups The various states fall into three groups: NeedsRemediation Status values in the NeedsRemediation group indicate that the package is in a bad or broken state and must be repaired before it can be used. NotAvailable Status values in the NotAvailable group indicate that Windows has the package installed, but it cannot be used at this time. ...
Building Agents for Teams: Turning conversations into outcomes
The Microsoft Teams platform mission is to build the best collaborative platform in the world. We want to make it easy for developers to build agents that participate seamlessly in chats, channels, and meetings, helping teams turn ideas into action in real time. By bringing agents directly into the places where collaboration happens, we can help ensure work is completed in the moment rather than getting lost in a stream of conversation. Everything we build is designed to help developers create collaborative experiences that empower people and agents to work together more effectively. Today we're launching ...
Microspeak: Double-click and drill down
Please, tell me more.
How a Large Enterprise Migrated 47 TB and 7,000+ Collections to Azure DocumentDB in Under 48 Hours
Large database migrations are where good planning either pays off or falls apart. In a MongoDB-compatible estate with thousands of collections, the hard part is not only moving data. It is deciding what to move, what to defer, how to sequence the work, and how to keep the system stable while ingestion is running at full speed. That is exactly what we saw in a recent large enterprise migration to Azure DocumentDB: 47 TB of data, 7,000+ collections, a 27-node source footprint, and a production cutover window of less than 48 hours. The target was intentionally smaller and simpler: a 3-node, 16 TB Azure DocumentDB...
Pure Virtual C++ 2026 [Meet the Speakers, Part 2]: The AI-Native C++ Developer Workflow
Pure Virtual C++ 2026 streams Tuesday, July 21, 2026 at 9:00 AM PT, a free, one-day virtual conference for the whole C++ community, live on YouTube (Microsoft Reactor) and Twitch, with on-demand recordings on the Visual Studio YouTube channel afterward. Register now → Welcome to part two of our Meet the Speakers series. In part one we looked at building and running faster. Today we turn to the theme shaping the future of development: the AI-native C++ developer workflow. Why this theme matters C++ development is evolving into an AI-native, agent-driven experience and it's happening everywhere you work, f...
The Visual Studio Dev/Test Benefit: Freedom to Build, Test, and Experiment in Azure
One of the best parts of a Visual Studio Subscription is discovering benefits that can make your day-to-day development work easier and more cost effective. One benefit that deserves more attention is Azure Dev/Test pricing. In fact, it's one of the most valuable benefits included with a Visual Studio Subscription. If you're building cloud applications, testing new ideas, or maintaining multiple development environments, cloud costs can sometimes influence how often you experiment or how closely your development environment matches production. Azure Dev/Test pricing helps remove that barrier by giving eli...
Why don’t we just make the entire stack out of guard pages?
Guard pages all the way down?
Go 1.26.5-2 Microsoft build now available
A new release of the Microsoft build of Go is now available for download.
The case of the mysterious changes to integers when there shouldn’t have been any code generation effect
Decoding where those integer came from.
Shrinking Azure Pipeline task extensions using esbuild
TL;DR We bundled an internal Azure Pipelines task extension into a single bundled JavaScript file using esbuild. The task package dropped from tens of megabytes and thousands of files to three files per task ( , , and ). The change took about 20 lines of build tooling. We measured the payoff across our production pipelines: Spending less time downloading and extracting tasks means we can make more efficient use of our build infrastructure. If you publish a node-based Azure DevOps task extension that ships a large folder or thousands of small files, you can almost certainly benefit from this same chan...
Microsoft Agent Framework for Go public preview
Microsoft Agent Framework for Go is entering public preview, bringing Agent Framework concepts to Go developers building agents and multi-agent workflows.
Pure Virtual C++ 2026 [Meet the Speakers, Part 1]: Build Faster, Run Faster
Meet the speakers behind Pure Virtual C++ 2026. In part one, Ryan Shepherd and David Li show how to build faster and smaller with C++20 modules and GitHub Copilot build performance.
Modernize .NET applications in the GitHub Copilot app
Modernize .NET applications in the GitHub Copilot app. Follow your upgrade from assessment through execution in an interactive upgrade canvas.
SharePoint Copilot Apps Now in Public Preview: From Intent to Action in Microsoft 365 Copilot
SharePoint Copilot Apps are now in public preview, introducing a new way to bring guided, action-oriented business experiences into Microsoft 365 Copilot. By combining natural language reasoning with structured UX, validation, permissions, and deterministic operations, organizations can help users move from intent to trusted outcomes directly in the Copilot flow.
I’ve decoded a #pragma detect_mismatch error and fixed the mismatch, but I still get the error
You need to rebuild everything that was dependent on the change.
Agent Harness: Scaling the claw or harness capabilities
Part 3 of Build your own claw and harness with Microsoft Agent Framework. In Part 2 our personal finance assistant learned to work with your data safely: it reads your portfolio, asks before it trades, and remembers what matters across sessions. It's useful - but everything it knows is baked into one prompt, it does its work one step at a time, and it can't reach past the file-access tools to actually reorganize anything. This part makes the claw more capable along four axes: As before, we only supply what makes our agent ours; the harness provides the machinery. Let's take them in turn...
What’s New in vcpkg (June 2026)
These June updates add a vcpkg-tool switch to skip installation when packages are already cached, OpenHarmony (OHOS) platform support, and major library updates including curl 8.21, CGAL 6.2, and Graphviz 15.
From Noisy Queries to Precise Frames: Query Decomposition for Media Asset Search
How query decomposition separates metadata filters from visual intent to significantly improve media asset retrieval quality.
Go 1.26.5-1 and 1.25.12-1 Microsoft builds now available
A new release of the Microsoft build of Go including security fixes is now available for download.
Let’s Learn GitHub Copilot App – Free Virtual Training Event
Join us for a free online event series kicking off July 16 to learn how to get started with the GitHub Copilot App!
Announcing TypeScript 7.0
Today we are proud to announce the availability of TypeScript 7, a 10x faster native port of TypeScript! Since its early days, TypeScript has promised to deliver on JavaScript that scales. By bringing strong type-checking and rich tooling to the world of JavaScript, TypeScript made it possible to build non-trivial high-quality apps across platforms. Last year, our team unveiled TypeScript's next step in scaling: making every part of the toolset an order of magnitude faster. The mission was a native port of TypeScript built in Go that could make the most of modern hardware. This port was done as faithfull...
Building on Vercel’s eve + Azure Cosmos DB: An Agent That Remembers
Most "AI agent" demos forget everything the moment the process exits. That's fine for a toy project, but useless for anything real. An agent that helps you write, triage, or support needs two things a language model alone can't give it: durable state and the ability to recall the right context by meaning. This post shows how to build exactly that by integrating two pieces that fit together surprisingly well: Eve — Vercel's filesystem-first agent platform. Drop a file in agent/tools/, and it becomes a tool the model can call. Azure Cosmos DB JavaScript SDK — the official, promise-based client for Cosmos DB N...
The other kind of control flow guard check: The combined validate and call
A two-in-one package.
The hidden variables in your agent eval
This is the seventh article in a series about Agent Experience (AX): the practice of making AI coding agents work correctly with your technology. The series covers what you can and can't control in the agent stack, how to measure whether your extensions are helping or hurting, and how to iterate toward better outcomes. You build an eval. You run it on your machine. You get a score. Your colleague runs the same eval on their machine and gets a different score. Same scenario, same setup. What changed? In the previous article, we covered why public benchmarks can't tell you which model works best for your stack. T...
Agent Framework’s Orchestration Patterns Reach 1.0
Python's package is now 1.0.0. That puts Microsoft Agent Framework's orchestration layer at 1.0 across Python and .NET. Sequential, concurrent, group chat, handoff, and magentic orchestration are now stable in both SDKs. You can pick the coordination pattern that fits your problem instead of choosing around SDK maturity. Magentic is the best example of why that matters. It is the least hand-wired pattern: you give it a goal, a manager, and a set of specialists, then the manager decides how the team should work. At the workflow layer, Agent Framework lets you build the graph yourself. Executors do the wo...
Applications Are Not Packages
Applications are not packages. Let me repeat. Applications are not packages. This is a common misconception due to the (too) often blurring of concepts and terminology. Packages and applications are different and distinct things, so it's important to understand what those words mean (to MSIX). Packages and Applications Are Different Things An application is distributed via a package. That's not the same as being the package. A package is a unit of distribution and deployment. You can install a package, or update a package, or remove a package. You can't install half a package, or remove part of a package. A...
How did Windows 95 decide that a setup program ran?
It used some heuristics.
Don’t rewrite your CLI for agents
There's advice making the rounds: replace your CLI args with a single payload so agents can use your tool more effectively. The thinking being, that agents already think in structured formats, and nested data maps cleanly to JSON. Flat args on the other hand, force awkward conventions like repeating to delimit multi-value groups, which is inherently ambiguous. Not to mention, that the agent needs to get the types of all values right. It's a reasonable hypothesis, and we wanted to know if it holds up under measurement. The data we collected, showed something interesting. What we tested We built a synthetic CL...
Agent Skills for .NET Is Now Released
You can now give your .NET agents reusable packages of domain expertise - instructions, reference documents, and scripts they load only when a task needs them - through a stable, production-ready API. Agent Skills for .NET in Microsoft Agent Framework has moved out of experimental preview - the attribute is removed and the API is stable. Teams can build skills, ship them independently, and combine them in any agent, with the governance controls enterprises need before putting agents into production. If you've been following our earlier posts on file-based skills and authoring modes with script execution, ever...
What’s New in Microsoft Foundry | June 2026
Claude is now generally available in Microsoft Foundry. Here's everything else that shipped between Build 2026 and the end of June — autopilot agents, expanded Toolboxes and Routines, Agent Optimizer's private preview, and more.
See our new Azure Cosmos DB Design Patterns
Design patterns are where good data modeling lives or dies. In a NoSQL database like Azure Cosmos DB, the difference between a schema that scales to millions of operations per second and one that fights you at every turn usually comes down to a handful of well‑understood patterns: how you partition, how you version, how you fan out work, how you keep concurrent writers from stepping on each other. That's exactly what the Azure Cosmos DB Design Patterns repo is for. It's a growing, hands‑on collection of small, focused samples — each one isolates a single pattern, explains why it matters, and lets you run it your...
I opened a file with FILE_FLAG_DELETE_ON_CLOSE , but now I changed my mind
You can't change your mind, but you can do it a different way.
Need a different partition key in Azure Cosmos DB? Pick the right approach
Once you create a container, its partition key is fixed at creation, and you can’t change it in place. However, if your original key starts causing problems like cross-partition queries or hot partitions, you need to consider your options for changing it. This post explains the mechanics of changing a partition key, and the tradeoffs between the options. First, think about the intent of your change: Note that every “change the partition key” option is actually a move: get your data into a new container that has the new key, then point your app at it. The exception is a Global Secondary Index, whic...
Not all model upgrades are upgrades
A new model drops with lower per-token pricing and better benchmarks. You switch. A week later someone asks why the agent is burning 12x more tokens on the same task while producing worse output. We ran 150 agent tasks across 15 scenarios on two models, Claude Sonnet 4.6 and Claude Sonnet 5, using GitHub Copilot Chat in VS Code on Windows. The scenarios covered two types of work: architecture and design tasks grounded in Microsoft Learn documentation, and SharePoint Framework project upgrades. Sonnet 5 is the newer model with 33% lower per-token pricing across every token category. The assumption we wanted to te...
Azure SDK Release (June 2026)
Azure SDK releases every month. In this post, you'll find this month's highlights and release notes.
How did we conclude that CcNamespace.dll was the ringleader of a group of DLLs that unloaded prematurely?
Contextual clues.
Enabling MLflow OpenAI Autolog on PySpark Workers
When distributing LLM calls across PySpark workers via mapInPandas, MLflow autolog silently fails. Here is how to fix it.
Pure Virtual C++ 2026 Talks Announced
The featured talks for Pure Virtual C++ 2026 are here! Join us on July 21 for a free, one-day virtual conference for the whole C++ community.
The case of the thread executing from an unloaded third-party DLL
Oops, I didn't realize that I was still doing that.
Fundamentals of Azure DevOps with SQL projects
Building automated pipelines with your SQL database projects enables you to build a rich CI/CD ecosystem to ensure that your application is being deployed with good quality code and at high confidence of success. SQL Database Projects are compatible with just about every automation environment because fundamentally they're built on top of the .NET SDK, a free and cross-platform development platform. You can develop the Microsoft.Build.Sql projects in VS Code and SQL Server Management Studio (SSMS). In this post, we'll take a look at the things you need to know to get started with building and deploying SQL projec...
Upcoming Change: NTLM Removal in Git (libcurl) – Impact to Azure DevOps Server Customers
Overview In September 2026, NTLM support will be removed from libcurl, which is used by Git for HTTP(S) operations. As a result, Git operations over HTTPS against Azure DevOps Server (on-premises) will stop working for customers who rely on NTLM authentication. This change is part of a broader industry move toward more secure authentication mechanisms. Many environments may be affected even if they believe they are using Kerberos. This is because Negotiate (SPNEGO) authentication can silently fall back to NTLM when Kerberos is not properly configured, leading to unintentional dependency on NTLM. If your enviro...
Audit Frontier AI Agents with SQL MCP Server
With On-Behalf-Of authentication, SQL MCP Server lets agents access Microsoft SQL without losing the user identity behind the request. Azure SQL can audit the signed-in user who invoked the operation, not just the agent, app, or MCP server that carried it out.
What AI benchmarks are not telling you
This is the sixth article in a series about Agent Experience (AX): the practice of making AI coding agents work correctly with your technology. The series covers what you can and can't control in the agent stack, how to measure whether your extensions are helping or hurting, and how to iterate toward better outcomes. We love benchmarks. A new model drops, the leaderboard says 92% on SWE-bench, and your timeline declares it "the best coding model." You switch to it, run your agent on your codebase, and outcomes are... the same. Maybe worse. The leaderboard said 92%, so what happened? In the previous article, we ...
It rather involved being on the other side of this airtight hatchway: Changing administrative settings
Unlocking the door from the inside.
What’s new across Microsoft SQL in 2026 so far (SQL Server, Azure SQL, and SQL database in Fabric)
We’re halfway through 2026, and Microsoft SQL has not slowed down. Since SQLCon/FabCon in March (where we released a ton of things, and those updates can be found in this updates video), we shipped a wave of updates across SQL Server, Azure SQL, and SQL database in Fabric, with Microsoft Build 2026 as the centerpiece. If you want the details on Build, start with my recap blog, The Era of the Agentic Database Developer. This post collects everything new from mid-March through today, organized by service so you can find what matters to you. Every item is tagged Preview or Generally Available with a link. You can...
Access GitHub Copilot in JetBrains – GitHub Copilot Is Now a Native Agent in JetBrains AI Assistant
GitHub Copilot is now a first-class, native option in JetBrains AI Assistant agent picker(read more). You can select it directly from the chat interface, no extra configuration needed, and immediately put Copilot to work on real coding tasks without ever leaving AI Assistant. Millions of developers already rely on the GitHub Copilot plugin as their AI pair programmer inside JetBrains IDEs. Copilot has long been reachable through JetBrains AI Assistant via the Agent Client Protocol (ACP), but today we are taking that integration a significant step further. This is not a separate window or a plugin ...
SharePoint Framework (SPFx) roadmap update – July 2026
June celebrates the announcement of upcoming SharePoint Copilot Apps, ships the quality-focused SPFx 1.23.2 release, and shares the roadmap for the AI era - shaped by your feedback.