Document GitHub Copilot Coding Agent capabilities vs standard Copilot

[Copilot]

User asked "What can you do that GitHub Copilot doesn't already do?" - created documentation explaining the differences between standard GitHub Copilot (IDE autocomplete/chat) and the autonomous Coding Agent.

Changes

• New file: VSCode/instructions/GitHub Copilot Agent Capabilities.md
  • 11 capability comparisons (autonomous execution, repository-wide analysis, CI/CD integration, security scanning, multi-file coordination, etc.)
  • Practical example: implementing user authentication end-to-end
  • Guidance on when to use each tool

Key Distinctions

Standard Copilot: Suggests code, requires manual implementation/testing/validation

Coding Agent: Explores repo → plans implementation → edits files → runs builds/tests → debugs failures → scans for vulnerabilities → commits working solution

File placed in VSCode/instructions/ alongside existing coding guidelines (General Coding Guidelines.instructions.md, PowerShell Style Guidelines.instructions.md, etc.)

Original prompt

What can you do that GitHub Copilot doesn't already do?

The user has attached the following file paths as relevant context:

• VSCode\instructions\General Coding Guidelines.instructions.md

Created from VS Code via the GitHub Pull Request extension.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.