AI Coding Agents

A coding agent does more than autocomplete. It reads the whole repository, edits several files, runs tests and commands, and keeps iterating on the results. Some are official tools from model labs, built around their own models; others are open-source harnesses that work with any model. This list covers the terminal, editor and cloud agents developers use most today.

How to choose

  • Model lock-in: does it only run the vendor's own models (Claude Code, Codex), or any model including local ones (OpenCode, Pi, Aider)?
  • Pricing: included in a subscription, pay-as-you-go API usage, or a free tier — token use on long tasks varies widely
  • How it works: terminal CLI, editor extension or asynchronous cloud runs, and whether it supports subagents, plan mode and headless runs in CI
  • Extensibility: MCP, skills, hooks, plugins and project instruction files such as AGENTS.md
  • Permissions and safety: does it ask before running commands, is there a sandbox, and is it safe to run unattended on a production repository?

Recommended software

19 in total, ranked by community recommendations, ratings and likes

Claude Code is an AI coding assistant from Anthropic. It works as an agent in your command-line terminal (CLI) and helps developers complete complex programming tasks through natural-language instructions.

A software development tool from OpenAI.

DeepSeek Harness (command-line name: dsh) is an open-source agent harness developed by DeepSeek AI. It uses an "everything is a plugin" architecture and is built on Cordis. It is currently in developer preview, and you can launch the Web UI with npx.

AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

The open source coding agent.

An integrated development environment from Google with added AI support

GitHub Copilot CLI is GitHub's AI coding agent for the terminal. It can read, write, and run code from the command line. It connects directly to GitHub issues and pull requests, and can take you all the way from planning to merging your code.

Cursor is a proprietary AI-assisted IDE built on Visual Studio Code. With a deep understanding of your project's context, it generates, refactors, debugs and documents code, significantly improving development speed and code quality.

Grok Build is a terminal coding agent and command-line tool from xAI, currently in beta. It can plan tasks, edit multiple files, and run commands in your terminal to handle complex coding work.

A coding assistant from Kimi.

Qwen Code is a command-line AI coding assistant from the Qwen team (Tongyi Qianwen). In your terminal, you can use natural language to have it understand a codebase, generate and modify code, and carry out development tasks.

aider is AI pair programming in your terminal

Autonomous coding agent as an SDK, IDE extension, or CLI assistant.

Kilo Code is an open-source AI coding agent that works in VS Code, JetBrains, the command line, a desktop app and the cloud. It supports 500+ models, bills at cost with no markup, and also lets you use your own keys or local models.

an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

Glamourous agentic coding for all 💘

Amp is a coding agent and development environment built for frontier models. It can run unattended on remote orb machines in the cloud, or locally on your own machine through the CLI.

Factory Droid is Factory's terminal AI coding agent. Through an interactive terminal UI (TUI), you can delegate an entire ticket, refactor, or investigation to Droid. It supports slash commands, resumable sessions and missions, as well as Skills, MCP, hooks, and plugins.

🙌 OpenHands: AI-Driven Development

Other use cases