#github

GitHub Copilot makes local sandboxing generally available

GitHub has made local sandboxing generally available for GitHub Copilot, isolating command execution with granular controls over files, network access, credentials, and system capabilities.

GitHub Copilot to add intelligent local model routing

GitHub announced that GitHub Copilot will soon support intelligent local model routing under Project HydraFusion, automatically directing suitable tasks to local models to help users save AI credits.

ClaudeDevs outlines Claude Code cloud sessions and releases workflow field guide

ClaudeDevs explained that Claude Code cloud sessions run on a fresh VM for each task, allowing users to start several tasks at once that continue running after closing their laptop. The team also published a field guide detailing seven suitable workflows and initial GitHub setup.

GitHub launches ReviewBench open benchmark to evaluate AI code reviewers

GitHub has released ReviewBench, an open benchmark designed to evaluate whether AI code review agents identify critical issues without adding noise, based on 219 pull requests across 19 languages.

GitHub Copilot app puts the diff, terminal, and browser side by side

In the GitHub Copilot app, the diff, terminal, and browser sit together so reviewers can check, run, and preview agent code without switching tabs. A beginner guide covers using that layout.

GPT-6.1 Sol is generally available in GitHub Copilot

GitHub says OpenAIDevs’ GPT-6.1 Sol is now generally available in GitHub Copilot for agentic coding and terminal workflows, including the Copilot app, CLI, and code.

Codex Security Cloud includes Daybreak Blue by default

OpenAI is upgrading Codex Security Cloud so access to cyber-capable models through Daybreak Blue is included by default. It scans full GitHub repositories, reviews new commits, and prepares fixes as a plugin in Codex desktop and web.

GitHub Copilot app can run agents in parallel on separate Git worktrees

In the GitHub Copilot app, agents can run in parallel, with each session using its own Git worktree and context so building, reviewing, and testing can happen at the same time.

GitHub Copilot app's rebuilt pull request view stays fast on extreme diffs

GitHub says the view remains fast and smooth for a pull request with 2,200 files, more than 1 million changed lines, and more than 400 inline comments.

Cline Desktop v0.0.34 adds worktrees, GitHub PR status, and parallel subagents

Parallel tasks each run on a separate branch under ~/.cline/worktrees/<id>. The release also shows a branch’s pull request link, merge status, and CI checks when gh is available, and can assign several tasks to subagents in one session.

GitHub will host beginner livestreams on the Copilot SDK starting September 24

GitHub will host beginner-friendly livestreams starting September 24 on building agentic applications with the GitHub Copilot SDK without implementing the agent loop. The series includes .NET and Java sessions, with sign-up in six languages.

Claude Opus 5.5 is now generally available in GitHub Copilot

AnthropicAI's Claude Opus 5.5 is generally available in GitHub Copilot for agentic coding, long-running agentic tasks, and knowledge work. GitHub's early testing found task resolution comparable to Claude Opus 5, with significantly fewer steps and tokens and fast error recovery on multistep tasks.