GitHub Copilot's agent stack: cross-repo projects, parallel sessions, desktop app and automations
0xkarasy · x · 2026-11-06
GitHub Copilot's site lays out its full AI coding agent capabilities: move from a single task to projects spanning repositories without losing context, and start Copilot sessions from issues or PRs with task and repo context pre-loaded.
Other highlights: run tasks in parallel with per-task sessions tracked in one view; long-running tasks continue across the desktop app, CLI, VS Code, github.com and GitHub Mobile; run reusable agent tasks on demand, on a schedule, or from triggers; review local changes for code quality and security issues before opening a PR; combine Copilot code review with team feedback and CI, delegating follow-ups to the Copilot cloud agent; and match each task to the right model. The desktop app is available for macOS (Silicon).
More from coding & agent
- Codex's computer use beats Claude Code on legacy enterprise software, devs say — kimmonismus · 2026-11-06
- Jensen Huang: Don't Call Them Agent Swarms, They're Agent Teams — shashib · 2026-11-06
- Google donates Agent Substrate, gVisor and OpenRL to CNCF; Agent Substrate hits 1.0 on GKE — HowDevelop · 2026-11-06
- Deep-dive report on parallel coding agents: worktree isolation doubles costs, merge conflicts dominate — weswinder · 2026-11-06
- OpenAI's Justin Uberti to reveal GPT-Live speech-to-speech architecture at VapiCon 2026 — juberti · 2026-11-06
- Approved action ran twice in 3 agent frameworks: crash-resume test — KaleidoscopeNew9259 · 2026-11-06