GitHub Copilot Chat

Processes natural language questions about code within a sidebar chat interface, leveraging the currently open file and project context to provide explanations, suggestions, and code analysis. The system maintains conversation history within a session and can reference multiple files in the workspace, enabling developers to ask follow-up questions about implementation details, architectural patterns, or debugging strategies without leaving the editor.

Triggered via Ctrl+I (Windows/Linux) or Cmd+I (macOS), this capability opens an inline editor within the current file where developers can describe desired code changes in natural language. The system generates code modifications, inserts them at the cursor position, and allows accept/reject workflows via Tab key acceptance or explicit dismissal. Operates on the current file context and understands surrounding code structure for coherent insertions.

Copilot can generate unit tests, integration tests, and test cases based on code analysis and developer requests. The system understands test frameworks (Jest, pytest, JUnit, etc.) and generates tests that cover common scenarios, edge cases, and error conditions. Tests are generated in the appropriate format for the project's test framework and can be validated by running them against the generated or existing code.

When developers encounter errors or bugs, they can describe the problem or paste error messages into the chat, and Copilot analyzes the error, identifies root causes, and generates fixes. The system understands stack traces, error messages, and code context to diagnose issues and suggest corrections. For autonomous agents, this integrates with test execution — when tests fail, agents analyze the failure and automatically generate fixes.

Copilot can refactor code to improve structure, readability, and adherence to design patterns. The system understands architectural patterns, design principles, and code smells, and can suggest refactorings that improve code quality without changing behavior. For multi-file refactoring, agents can update multiple files simultaneously while ensuring tests continue to pass, enabling large-scale architectural improvements.

Copilot Chat supports running multiple agent sessions in parallel, with a central session management UI that allows developers to track, switch between, and manage multiple concurrent tasks. Each session maintains its own conversation history and execution context, enabling developers to work on multiple features or refactoring tasks simultaneously without context loss. Sessions can be paused, resumed, or terminated independently.

Copilot CLI enables running agents in the background outside of VS Code, allowing long-running tasks (like multi-file refactoring or feature implementation) to execute without blocking the editor. Results can be reviewed and integrated back into the project, enabling developers to continue editing while agents work asynchronously. This decouples agent execution from the IDE, enabling more flexible workflows.

Provides real-time inline code suggestions as developers type, displaying predicted code completions in light gray text that can be accepted with Tab key. The system learns from context (current file, surrounding code, project patterns) to predict not just the next line but the next logical edit, enabling developers to accept multi-line suggestions or dismiss and continue typing. Operates continuously without explicit invocation.

Copilot agents can implement features end-to-end across multiple files, executing terminal commands to run tests, and automatically correcting code when tests fail. The agent decomposes tasks into steps, generates code in multiple files simultaneously, runs the test suite, analyzes failures, and iterates until tests pass. This operates asynchronously and can run in the background via Copilot CLI, with central session management for parallel agent execution.

Copilot agents can create Git branches, commit code changes, and open pull requests directly from the chat interface, enabling end-to-end feature implementation without manual Git operations. The agent handles branch naming, commit messages, and PR creation, integrating with GitHub's API to manage the full workflow. This capability bridges code generation and version control, allowing developers to request features and receive them as ready-to-review PRs.

Developers can define custom agents with specialized capabilities, custom instructions, and project-specific context. Agents can be configured with particular skills (e.g., 'database expert', 'frontend specialist'), given access to custom tools via MCP servers or extensions, and instructed with project-wide coding guidelines. This enables teams to create domain-specific assistants that understand project conventions and can be reused across multiple development sessions.

Copilot agents can be extended with external tools via Model Context Protocol (MCP) servers or VS Code extensions, enabling access to specialized APIs, data sources, and services. Agents can invoke these tools as part of their reasoning and code generation, treating external services as callable functions within their action space. This architecture allows integration with databases, internal APIs, monitoring systems, or any service that exposes an MCP interface.

Developers can define project-wide custom instructions that are automatically injected into every agent interaction, specifying coding guidelines, architectural patterns, naming conventions, and task-specific context. These instructions are stored at the project level and applied consistently across all chat sessions and agents, ensuring that all AI-assisted development adheres to project standards without per-request manual context.

A specialized agent persona focused on breaking down complex development tasks into smaller, actionable steps. The planning agent analyzes feature requests or refactoring goals, identifies dependencies, suggests implementation order, and can hand off to other agents for execution. This capability enables developers to request high-level goals and receive structured plans that can be executed incrementally or delegated to specialized agents.

When a developer's request is ambiguous or lacks sufficient context, agents can generate clarifying questions to gather missing information before proceeding with code generation. This capability prevents wasted effort on incorrect implementations by ensuring agents understand requirements before committing to a solution. Questions are presented in the chat interface for the developer to answer, refining the request iteratively.