AI Coding Agents and Assistants

AI coding agents and assistants help developers write, review, explain, test, and change code. Some work as IDE extensions, while others run in the terminal or in a cloud workspace and can take multi-step actions across a repository.

If you want a full AI-first editor, start with AI Native Editors. This page focuses on assistants and agents that can work with an existing development workflow.

Tool Options

GitHub Copilot AI assistance inside familiar IDEs and GitHub workflows, including chat, completions, and agent features.

Claude Code Anthropic's terminal coding agent for inspecting repositories, editing files, and running development tasks.

OpenAI Codex OpenAI coding workflows for agentic implementation, review, and repository-aware development.

Gemini CLI Google Gemini assistance from the command line for developer workflows and repository tasks.

Tool	Made By	Type	Where It Runs
GitHub Copilot	Microsoft / OpenAI	IDE assistant and coding agent	VS Code, Visual Studio, JetBrains, GitHub
Claude Code	Anthropic	Agentic CLI	Terminal
OpenAI Codex	OpenAI	Agentic coding assistant	Terminal / cloud workflows
Gemini CLI	Google	CLI assistant	Terminal

How to Choose

If you want…	Start With
AI help inside a familiar IDE	GitHub Copilot
A terminal agent that can inspect and edit a repo	Claude Code
OpenAI-based coding workflows	OpenAI Codex
Gemini-based command-line assistance	Gemini CLI

What These Tools Usually Do

Read nearby files or wider project context.
Suggest code, tests, and refactors.
Explain unfamiliar code.
Help debug compiler, test, and runtime errors.
In agentic workflows, plan and apply multi-file changes.

For IDE-first tools such as Cursor, Windsurf, Kiro, and Google Anti-Gravity, use the AI Native Editors page.