Copilot Workspace
A GitHub Next research project exploring task-centric AI development environments where developers describe a goal and an AI agent plans and implements the solution.
At a Glance
Free access to the Copilot Workspace research prototype from GitHub Next.
Engagement
Available On
Alternatives
Updated May 2026
About Copilot Workspace
Copilot Workspace is a research project from GitHub Next, the experimental research and engineering team at GitHub. It explores a new paradigm for software development where developers start with a task or issue rather than a blank editor, and an AI system generates a plan, proposes code changes, and iterates toward a solution. The project sits within GitHub Next's broader investigation into the future of software development tooling.
What It Is
Copilot Workspace is a task-centric development environment prototype built by GitHub Next. Rather than augmenting a traditional code editor with AI suggestions, it reimagines the development workflow around a goal: a developer describes what they want to accomplish — often starting from a GitHub issue — and the system produces a step-by-step plan, generates the necessary code edits across files, and allows the developer to review, adjust, and iterate. The core idea is to shift the unit of work from "write code" to "accomplish a task."
GitHub Next Context
GitHub Next describes itself as a team of researchers and engineers at GitHub exploring things "beyond the adjacent possible." The team prototypes tools and technologies intended to change software development craft, and Copilot Workspace is one of its higher-profile projects. Other active GitHub Next projects include Repo Mind (global codebase understanding for humans and AI agents), Autoloop (goal-driven agentic loops), Agentic Workflows (natural-language programming for GitHub Actions), and Discovery Agent (agentic setup, build, and testing of repositories). GitHub Next publishes research findings, posts, and project updates publicly at githubnext.com.
Agentic Development Approach
Copilot Workspace represents GitHub Next's exploration of agentic software development — where AI systems take multi-step actions toward a goal rather than providing single-turn completions. This connects to the team's broader research themes visible across their 2025–2026 project portfolio: automated repository maintenance, continuous AI in platform-based collaboration, and formal verification with near-zero human labor. GitHub Next's published research on repo-assist, for example, documents deploying a proactive AI repository agent across 13 open source repositories, with the team reporting 578 issues closed and significant increases in issue closure and PR merge velocity.
Current Status
Copilot Workspace is listed as a project on the GitHub Next projects page. GitHub Next projects carry status labels such as "Research Prototype," "WIP," "Napkin Sketch," or "Completed." The project page at githubnext.com/projects/copilot-workspace is the canonical source for current availability and status. GitHub Next regularly publishes posts and updates about its projects, and the team's site was refreshed in early 2026 to better surface intermediate research artifacts and working notes alongside finished project pages.
Who It Is For
Copilot Workspace targets software developers who want to explore AI-assisted, task-driven development workflows. As a GitHub Next research prototype, it is primarily of interest to developers curious about the frontier of AI coding tools, researchers studying human-AI collaboration in software engineering, and early adopters willing to engage with experimental tooling. GitHub Next's work is conducted largely in the open, with findings, prototypes, and research reports shared publicly.
Community Discussions
Be the first to start a conversation about Copilot Workspace
Share your experience with Copilot Workspace, ask questions, or help others learn from your insights.
Pricing
Research Prototype
Free access to the Copilot Workspace research prototype from GitHub Next.
- Task-centric AI development environment
- AI-generated plans from GitHub issues
- Multi-file code change proposals
- Iterative review and adjustment
Capabilities
Key Features
- Task-centric development environment
- AI-generated step-by-step plans from issues or goals
- Multi-file code change proposals
- Iterative review and adjustment workflow
- GitHub issue integration
- Agentic code generation
