The open platform for AI agents
Run any agent with any model across your full project in an open, extensible platform.
Download
Pick your agent experience
Run agents locally, in the cloud, from the CLI, or bring your own. Use what fits your workflow.
Work interactively with agents in your project
Let the agent run inside VS Code with access to your codebase and tools. See changes applied directly and guide the agent when you need to.
Best for: iterative tasks where you want to stay in the loop and course-correct quickly and often: planning, prototyping, front-end development, or debugging.
Learn more about local agents
Delegate tasks to agents in the cloud
Assign a task and move on. Cloud agents continue working in the background, implement changes, and open a pull request. Your team reviews, iterates, and merges without waiting on each other.
Best for: async workflows or tasks that require team collaboration.
Explore cloud agents
Run agents in an isolated environment on your machine
Agents work in their own workspace on your local machine without interfering with your project. Assign a task, keep working, and check the results when you're ready.
Best for: async workflows or tasks that don't require constant supervision and interaction: test generation, migrations, code refactoring.
Learn about Copilot CLI
Use third-party agents directly in your project
Run agents from Anthropic, OpenAI, and others locally or in the cloud. Use the commands and capabilities each agent provides, right alongside the tools you already know.
Best for: teams with existing agent preferences, use your favorite agent harness without giving up VS Code's editing power.
Explore third-party agentsAll your agents, one view
Track every agent session from a single place, across Copilot and third-party agents. See what's running, what needs your attention, and pick up any session where it left off.
Manage agent sessions
Swap the model, keep the workflow
Choose from dozens of models across OpenAI, Anthropic, Google, and more. Bring your own API key to use any model from any provider, including models you host yourself.
Learn more about language models
Agents that work your way
Define custom instructions, connect MCP servers, and add skills so agents follow your team's conventions and workflows. Browse the marketplace to install ready-made customizations from the community.
Learn how to customize your agents
Real tasks. Real results.
Agents help everyone on the team, from prototyping ideas to shipping features as pull requests.
Build a feature end-to-end
Describe a feature in natural language. The agent scaffolds, implements across files, and runs tests to verify.
Debug and fix failing tests
Point an agent at a failing test. It traces the root cause, applies a fix, and re-runs to confirm.
Refactor or migrate a codebase
Plan a migration between frameworks or APIs. The agent applies coordinated changes and verifies with builds.
Ship prototypes, not specs
Product managers and designers turn user feedback, a Figma file, or a rough idea into a working pull request the team can review and iterate on.
Audit accessibility and UX
An agent launches your app with Playwright, navigates to the feature, and verifies behavior, layout, and accessibility with screenshots.
Write docs, release notes, and summaries
Summarize the day's commits across repos into API docs, changelogs, release notes, or social posts.
And much more...
Deeper capabilities that make agents production-ready.
Planning
Agents research your codebase and produce a structured implementation plan before writing any code.
Agent memory
Agents remember your preferences, codebase conventions, and session context across conversations.
Subagents
Delegate focused tasks to context-isolated subagents that run in parallel and report back.
Agent tools
Built-in tools, MCP servers, and extensions give agents access to your development environment and external services.
Trust & safety
Run agents in sandboxed environments and configure auto-approvals for tool usage and file edits.
Enterprise policies
Manage agent capabilities and tool access across your organization with centralized policies.