What shipped & dropped across AI labs, today.
Today
Wed · Oct 28 · 20 posts
Hill-Climbing Agent Quality with Agents CLI Evaluation
This post walks through the Agents CLI evaluation loop for systematically improving agent performance through continuous testing and grading. It demonstrates how to catch prompt regressions, evaluate model backend swaps using side-by-side trace comparisons, and automate evaluation passes. Developers will also learn how to integrate nightly scheduled runs with threshold-based alerting to track quality over time.
GitHub Copilot SDK is becoming Microsoft Copilot SDK
We’re transitioning GitHub Copilot SDK to Microsoft Copilot SDK. What began as the SDK for the GitHub Copilot harness is becoming the programmable interface for the Microsoft Copilot harness, the…
Copilot code review: Merge with confidence with Max and approvals
GitHub Copilot code review now offers its most thorough review and can approve your pull request when it’s ready to merge. The new Max review effort level is now in…
Agent Merge watches and fixes pull requests in the cloud
With Agent Merge, Copilot watches your pull request and automatically fixes blockers, (e.g., failing Actions workflows and merge conflicts) that stand in the way of getting your pull request merged.…
Create and manage Copilot Automations in Visual Studio Code
Copilot Automations in Visual Studio Code can now run in the cloud. You can create and manage cloud automations from the Agents window without leaving your editor. With Copilot Automations,…
Copilot Automations now learn from outcomes and self-improve
Copilot Automations can now learn from their outcomes and propose improvements to themselves, helping them produce better results over time. It’s easy to create an automation, but harder to build…
New tools, triggers, and cost controls for Copilot Automations
With Copilot Automations, you can combine prompts, triggers, and tools to design workflows around the way your team works. Triggers define when your automation runs (e.g., “every day at 9am”…
Issue artifacts are now in public preview
Issue artifacts attach agent-generated outputs and other important context directly to an issue. By keeping plans, research, links, and visualizations alongside the work, they give coding agents richer context and…
GitHub Copilot in VS Code, Universe releases
This year, we focused on giving you more choice in how you work with GitHub Copilot in Visual Studio Code: which models you use, where your agents run, and how…
Work from anywhere with a new, connected Copilot cloud experience
Copilot is now one connected experience in the cloud. Start a cloud task wherever you are, on github.com, GitHub Mobile, the Copilot CLI, the Copilot app, VS Code, or right…
GitHub Copilot in JetBrains Universe releases
Today, we’re making GitHub Copilot in JetBrains more capable, secure, and seamless. From GitHub Copilot Fusion’s smarter model orchestration to local sandboxes and next edit suggestions (NES), this update brings…
Connect your tools to Copilot without all the setup
The context your agent needs is often outside your repository: a thread in Slack, a design in Figma, an error in Sentry. Until now, giving Copilot that context meant finding…
GitHub Agentic Workflows generally available
GitHub Agentic Workflows, an open source project from GitHub Next, is now generally available. With agentic workflows, you can automate reasoning-based tasks like issue triage, CI failure analysis, documentation updates,…
GitHub Copilot Fusion now supports selection of local models
Today we are shipping intelligent routing for local models into experimental preview. GitHub Copilot Fusion can now choose between supported on-device models and cloud models, enabling intelligent local model routing…
Generate and edit images directly in Copilot
You can now generate and edit images using natural language in GitHub Copilot. When you ask for an image, your selected reasoning model calls a dedicated image-generation tool powered by…
Build shared workflows for your team with Copilot Automations
Shared Copilot Automations are rolling out, enabling teams to build and maintain AI-powered workflows together. With Copilot Automations, you can combine prompts, triggers, and tools to design workflows around the…
GitHub intelligence puts your code, context, and history to work
Agents should understand more than your codebase—they should understand how your team works. Your preferences, repository conventions, skills, and custom instructions guide how changes should be made, while code in…
Security extension for the GitHub Copilot app
You can now work through your repository’s GitHub code scanning alerts without leaving the GitHub Copilot app. The new GitHub security extension, now in public preview, lets you find the…
GitHub Enterprise now includes access to GitHub Copilot
Enterprises working with a GitHub or Microsoft account team can now get Copilot included with GitHub Enterprise for every licensed user. Previously, you had to manage two things: how many…
GitHub Copilot app, Universe releases
Work doesn’t always fit in a single repository or session. A feature can span several repositories. A launch plan can pull in specs, designs, and assets without touching a repository…