GitHub Changelog · Oct 24, 2026 · 1:22 AM

Copilot in Visual Studio adds PR exploration via Git agent; BYOM now requires the new Agent (Preview)

In the update posted Oct 22, the Git agent in Visual Studio 2026 can summarize an attached pull request with links to its comments and files (via GitHub or Azure DevOps MCP servers), and NuGet Audit warnings get a fixer that has Copilot patch vulnerable packages. Bring Your Own Model now only works in the Copilot SDK-based Agent (Preview), so previously configured Ollama models must be re-added.

Read at github.blog

GitHub Copilot

AI coding assistance across editors, GitHub, and the CLI

More news

View all

Ollama 0.40.9 adds local text-to-speech with Phonon and an OpenAI-compatible /v1/audio/speech endpoint

Released Oct 22, ollama run phonon "..." speaks text aloud as it's generated (or pipe a chat model's reply into it), and apps can stream WAV or PCM audio from /v1/audio/speech using OpenAI's client. It runs on Apple Silicon only for now, English only, with a 4,096 character limit per request (docs)…

GitHub (ollama/ollama) 19 minutes ago

GitHub's new Teams app lets you @mention Copilot to turn an issue notification into a pull request

Announced Oct 23, the consolidated GitHub app for Microsoft Teams (formerly GitHub notifications) lets you @GitHub on a bug notification so Copilot investigates, fixes it, and opens a PR for your team to review. The legacy GitHub app becomes "GitHub (deprecated)" and retires on November 16, so Copilot in Teams users need to install the new one…

GitHub Changelog 2 hours ago

IBM makes Pipeline Automation for Z 1.2 generally available, pairing mainframe CI/CD with its Bob AI agent

GA as of Oct 23, version 1.2 bundles dependency based builds, ZCodeScan static analysis on the CLI, unit and functional testing, and Wazi Deploy into governed pipelines for z/OS apps. It's sold as a peer to IBM Bob Premium Package for Z, which supplies the AI assisted coding at each stage (product page)…

IBM Announcements 3 hours ago

Study: a coding agent's harness can triple cost per task without improving accuracy

Shandong University researchers held models fixed and ran Claude Code, OpenCode, and the 100 line mini SWE agent on 492 SWE bench Verified tasks, finding statistically equivalent success rates but up to 3x differences in cost per task. On the 45 hardest tasks, switching harnesses flipped results about as often as simply rerunning the same setup (13%), so small single run benchmark gaps say little…

VentureBeat 4 hours ago

GitHub makes Copilot's Agentic Autofix for code scanning alerts generally available

Now GA as of Oct 23, the Copilot coding agent investigates a CodeQL alert, writes a fix, reruns CodeQL to confirm it, and opens a draft PR, with up to 25 alerts batchable into one PR. It only covers alerts on the default branch and needs GitHub Code Security or Advanced Security on Enterprise Cloud, an admin enabled policy, and a Copilot billing method…

Visual Studio Magazine 5 hours ago

Spotted something we missed? Start a thread.