// category · 11 tools · 5 open source
AI Coding Agents
AI pair programmers, IDEs, terminal agents and app builders that write, refactor and ship code.
AI coding agents read your repository, edit files, run commands and tests, and loop until a change works, instead of only suggesting the next line. They come in three shapes, and most teams end up using more than one.
Terminal agents run in your shell and work on the folder you start them in. Claude Code, OpenAI Codex (the CLI and its cloud tasks), Gemini CLI, OpenCode and Aider belong here. They suit long, multistep jobs such as refactors, migrations and fixing a failing test suite. Our Claude Code vs Codex vs Gemini CLI comparison covers the big three, and best open source coding agents covers the ones you can inspect and self host.
Editor agents live inside an IDE. Cursor, GitHub Copilot, Cline and Zed combine inline completion with an agent mode that can plan and edit across files while you review each diff. Read best AI coding assistant and GitHub Copilot vs Claude Code if you are choosing one for a team.
App builders such as Replit and Lovable generate and host a whole app from a prompt. They are fastest for prototypes and internal tools, and they trade away some control over the code.
Pricing models differ more than features. Some agents are bundled with a subscription that has usage limits, some bill per token through an API key, and open source agents are free but still need a model. Google moved individual Gemini CLI users off the free tier in June 2026, which shows how quickly these terms change; check each vendor's pricing page before you commit.
Whatever you pick, two habits make agents more reliable. Write the project's rules into an instruction file (see our AGENTS.md guide), and give the agent reusable skills rather than repeating long prompts (see the agent skills guide). Review every change it proposes, and keep secrets out of the folders it can read. Official references: the Claude Code docs and the agents.md specification.
// guides
Guides to AI coding agents
What Is Agentic Coding? A Plain English Guide
What is agentic coding? An AI agent that reads your code, edits files, runs tests and repeats until the task is done. How it works, the tools, and how to start.
Best AI Coding Assistant in 2026
Pick the best AI coding assistant for your workflow: Copilot for GitHub teams, Cursor for an AI IDE, Claude Code for agentic refactors, plus free options.
AGENTS.md Guide: How to Write One (vs CLAUDE.md)
What AGENTS.md is, how Codex, Claude Code, Cursor, Copilot and Gemini CLI load it, how it compares with CLAUDE.md, plus a copy and paste template that works.
GitHub Copilot Alternative for Visual Studio (2026)
The best GitHub Copilot alternative for Visual Studio in 2026: Antigravity, JetBrains Junie, Claude Code, your own model key, Tabnine and CodeRush, checked Oct 27.
Cloudflare Security Audit Skill: How to Use It
How the Cloudflare security audit skill works, how to install it for Claude Code or Codex, what its six phases produce, and the sandbox rules it needs.
Best Pi Coding Agent Alternatives in 2026
Looking for a Pi coding agent alternative? Compare OpenCode, Claude Code, Codex CLI, Copilot CLI, Cline and Aider on plan mode, approvals and local models.
GitHub Copilot vs Claude Code: Key Differences
GitHub Copilot vs Claude Code: IDE autocomplete and GitHub agents versus a terminal first coding agent (pricing, models, MCP, and who should pick which).
Agent Skills Guide: How to Write a SKILL.md
Agent skills explained: how SKILL.md works, where Claude Code, Codex, Cursor, Copilot and Gemini CLI load skills, how they differ from MCP, plus a template.
Best Open Source Coding Agents in 2026
The best open source coding agents in 2026, compared on license, interface, model choice and upkeep: OpenCode, Cline, Codex CLI, Gemini CLI, goose and Aider.
Claude Code vs Codex vs Gemini CLI (2026)
Claude Code vs Codex vs Gemini CLI compared on pricing, open source, setup, MCP, instruction files and workflow, so you can pick the right terminal agent.
Power BI MCP Server: Which One to Use (2026)
Which Power BI MCP server to use in 2026: Fabric IQ for questions, the Authoring server (hosted or local) for model edits, with setup for Copilot and Claude.
Best Design Skills for Claude Code (2026)
Impeccable, Hallmark, Taste Skill and Anthropic's frontend design skill compared: what each one does, how to install it, and which to pick for your agent.
What Is MCP? The Model Context Protocol Explained
What is MCP? The Model Context Protocol explained: hosts, clients, servers, tools, resources, the stateless 2026 spec, security risks and a tested first run.
Is Gemini CLI Free for Pro Users? Explained
Is Gemini CLI free for Pro users? No. Google AI Pro access ended June 18, 2026. Here's what Pro and Ultra get in Antigravity CLI and what Gemini CLI costs now.
How to Run an MCP Server Locally (2026 Guide)
How to run an MCP server locally with npx, uvx or Docker, test it with the MCP Inspector, then connect it to Claude Code, Cursor, VS Code or Codex safely.
Claude Code Review: How to Set It Up and Use It
Claude Code Review is now generally available. How the GitHub service differs from /code-review, what it costs, how to tune it with REVIEW.md, and who can use it.
Claude Code Mods: What They Are, How to Build One
Claude Code mods are plugins whose JavaScript runs inside Claude Code to add panes, commands and tool call rules. How to install, build and vet one safely.
How to Use Gemini CLI in Terminal (2026 Guide)
How to use Gemini CLI in terminal: install it, sign in with a key that still works, run your first task, and learn @ files, ! shell, slash commands and MCP.
How to Use MCP Server in GitHub Copilot
How to use an MCP server in GitHub Copilot: the config file for VS Code, JetBrains, Visual Studio, Copilot CLI and the cloud agent, plus the GitHub MCP server.
LiteLLM Compromised: What to Do About 1.107.0
LiteLLM compromised again: release 1.107.0 on PyPI and Docker carried a credential stealer. How to check every host, what to rotate and how to pin safely.
How to Build an MCP Server in Python (2026 Guide)
Build a working MCP server in Python with the official SDK 2.x and the stateless 2026-07-28 spec, then connect it to Claude Code, Cursor, and VS Code hosts.
NVIDIA OpenShell: What It Is and How to Use It
NVIDIA OpenShell explained: how the open source agent sandbox enforces file, network and credential policy, how to install it, and how it compares to Docker.
OpenCode 2.0: What's New and How to Upgrade
OpenCode 2.0 adds a shared background service, Code Mode for MCP and a new plugin API, and 2.1 renames SDK fields. Here's what changed and how to upgrade.
LLM Gateway vs OpenRouter: Which to Use?
What an LLM gateway is, and when to self host one like LiteLLM instead of using OpenRouter: routing, fallbacks, caching, keys, fees and security in 2026.