Browse Artificial Intelligence News (415)

Jack Batzner shows how to add a governance layer to MCP-based AI agents in .NET using the Agent Governance Toolkit, including policy-driven tool-call authorization, security scanning of tool definitions, response sanitization to reduce prompt-injection risk, and built-in audit/telemetry via OpenTelemetry.

Sergey Menshykh announces A2A Protocol v1.0 support in Microsoft Agent Framework for .NET, showing how to discover and call remote A2A agents, stream responses, and host your own agents as A2A endpoints in ASP.NET Core with updated v1 hosting APIs and migration notes from v0.3.

Mark Downie covers the April Visual Studio 2026 update, focusing on GitHub Copilot’s new cloud agent workflow, user-level custom agents, and a Debugger Agent that validates fixes against real runtime behavior, plus improvements to C++ agent tools, IntelliSense vs Copilot completion priority, and configurable Copilot shortcuts.

Microsoft Defender Security Research Team explains how Microsoft Sentinel UEBA enriches AWS CloudTrail logs with simple true/false behavioral signals and built-in anomalies, helping detection engineers write simpler KQL, reduce false positives, and triage suspicious AWS activity faster.

stclarke shares a LinkedIn post highlighting a Microsoft Foundry demo of durable, stateful agents that persist across sessions, orchestrate tools and models, and use checkpointing plus evaluation to improve over long-running workflows.

Allison announces that GitHub Copilot Student is removing GPT-5.3-Codex from the model picker, while keeping it available via auto model selection. The post explains this as part of temporary reliability/performance measures and points to documentation on supported models and upcoming usage-based billing changes.

Allison shares a GitHub Changelog update: Copilot cloud agent now starts over 20% faster by using optimized runner environments prebuilt with GitHub Actions custom images, reducing environment startup overhead when Copilot begins work from issues, PRs, or the Agents tab.

Allison announces a billing change for GitHub Copilot code review: starting June 1, 2026, reviews will consume both Copilot AI Credits and (for private repos) GitHub Actions minutes, with guidance on checking usage, budgets, and runner configuration.

Mario Rodriguez announces that GitHub Copilot plans will move to usage-based billing on June 1, 2026, replacing premium request units with GitHub AI Credits based on token usage. The post explains what changes for individuals and organizations, including pooled credits, budget controls, and how Copilot code review also uses GitHub Actions minutes.

stclarke shares a LinkedIn post about Cricket Australia’s Live app, highlighting how Azure OpenAI and Azure Cosmos DB power “AI Insights” that let fans explore match context, player stats, and cricket history with fast, personalized responses.

Victor Colin Amador announces that Azure MCP Server can now be installed as an MCP Bundle (.mcpb), enabling a one-click, no-runtime setup in Claude Desktop and other MCP clients, plus a quick walkthrough of installation, Azure authentication, and what Azure tools become available after setup.

Allison announces that GPT-5.5 is rolling out to GitHub Copilot, highlighting improved performance on complex, multi-step agentic coding tasks, plus where you can select it (IDEs, CLI, web, mobile) and what admins need to enable for Business/Enterprise.

Allison announces updates to GitHub Copilot for JetBrains IDEs, including inline agent mode (public preview), improvements to Next Edit Suggestions, and new global/granular auto-approve controls for tool calls like terminal commands and file edits, plus UX and reliability refinements.

Wes Steyn breaks down the main chat history storage patterns for AI agents—service-managed vs client-managed—and explains how Microsoft Agent Framework uses AgentSession and pluggable history providers to switch between providers (including Azure OpenAI and Foundry) without rewriting your agent code.

The Visual Studio Code Team shares the VS Code 1.118 (Insiders) updates, focused on GitHub Copilot CLI improvements like session-title APIs, session switching keybindings, auto model selection, model badges in chat, and a Copilot CLI SDK change to resolve node-pty via hostRequire.

Naomi Moneypenny announces GPT-5.5 general availability in Microsoft Foundry and explains what’s new (agentic coding, long-context reasoning, token efficiency) plus how Foundry Agent Service helps run hosted agents with isolation, Entra identity, and governance for production use.

Allison announces improvements to GitHub Copilot Chat for pull requests, including richer PR context (comments, file changes, commits, reviews) plus new capabilities to review and summarize PRs directly from GitHub surfaces.

Allison announces an update to the GitHub Copilot usage metrics REST API: user-level reports now include a new `used_copilot_cloud_agent` boolean field (matching the Copilot cloud agent rename). The legacy `used_copilot_coding_agent` field remains for backward compatibility until August 1, 2026.

Yohan Lasorsa and Dan Wahlin introduce LangChain.js for Beginners, a free open-source course with 8 chapters and 70+ runnable TypeScript examples that teaches how to build agentic AI apps (tools, function calling, ReAct agents, MCP, and Agentic RAG) using JavaScript/TypeScript.

Allison announces an update to GitHub Copilot Chat on github.com that improves debugging when you paste a stack trace. Copilot now recognizes stack traces more reliably and responds with a more structured root-cause analysis, including evidence from code references, confidence level, and suggested fixes.