Awesome AI Coding Tools

A curated list of AI-powered tools for developers — editors, agents, code review, testing, CLI tools, MCP servers, and more.

AI coding has gone from autocomplete to autonomous teams in two years. This list helps you cut through the noise: the tools that ship, the tools that are interesting, and the tools that are quietly winning. Curated for working developers, not benchmark-chasers.

Maintained by LaunchApp. Contributions welcome — see contributing.md.

---

Contents

• AI Code Editors & IDEs
• In-Editor Assistants & Completion
• Autonomous Coding Agents
• CLI & Terminal Coding Tools
• Agent Orchestrators & Multi-Agent
• Code Review & PR Automation
• Testing & QA
• Code Search & Codebase Intelligence
• Documentation Generation
• Web-Based App Builders
• Open-Source & Self-Hosted
• Model Context Protocol (MCP)
• DevOps & CI/CD
• Models for Coding
• Prompt & Context Engineering
• Learning Resources
• Related Lists

---

Legend

• 🟢 Open source
• 💰 Paid / commercial
• 🆓 Free tier available
• 🏠 Self-hostable
• ⭐ Editor's pick

---

AI Code Editors & IDEs

Full IDEs and forks built around AI as a first-class feature.

• Cursor ⭐ 💰 — AI-first VS Code fork. Multi-file edits, agent mode, codebase chat. Currently the category leader.
• OpenMagic 🟢 — Browser-side AI coding toolbar for live web app edits with context capture and approved diffs.
• Windsurf 💰 — VS Code fork from Codeium with the "Cascade" agent. Strong long-running task support.
• Zed 🟢 🆓 — Rust-built editor with native AI assistant and multibuffer edits. Fast.
• Void 🟢 — Open-source Cursor alternative. Bring your own model.
• Trae 💰 — ByteDance's AI IDE with Builder/Chat modes.
• PearAI 🟢 — Open-source AI editor fork.
• Volt 🟢 🆓 — Open-source agentic workspace on the VS Code core that drives Claude Code, Codex and local Ollama models, with hunk-level review of agent edits and Git checkpoints.

In-Editor Assistants & Completion

Extensions that add AI to existing editors (VS Code, JetBrains, Vim, etc).

• GitHub Copilot ⭐ 💰 — The original. Inline completion, chat, agent mode, PR review.
• Continue 🟢 🆓 — Open-source autopilot for VS Code and JetBrains. Configurable models.
• Codeium 🆓 — Free AI completion across 70+ editors.
• Tabnine 💰 🏠 — Privacy-focused completion. On-prem deployment available.
• Supermaven 💰 🆓 — Fast, long-context completion. Acquired by Cursor.
• Cody 🆓 💰 — Sourcegraph's assistant with codebase-wide context.
• Amazon Q Developer 🆓 💰 — Formerly CodeWhisperer. Tight AWS integration.
• JetBrains AI Assistant 💰 — Native to all JetBrains IDEs.

Autonomous Coding Agents

Agents that take a goal and execute multi-step work — planning, editing, testing, iterating.

• Claude Code ⭐ 💰 — Anthropic's terminal-native agent. Strong tool use, MCP support.
• Aider 🟢 🆓 ⭐ — Pair-programs from your terminal with git-aware diffs. Model-agnostic.
• OpenAI Codex CLI 🟢 — Lightweight terminal agent from OpenAI.
• Gemini CLI 🟢 — Google's agent for the terminal.
• Harness Desktop 🟢 — Windows client for the official DeepSeek Harness coding workbench with themes, provider and subagent model routing, and an in-app plugin and Skills marketplace.
• Devin 💰 — Cognition's autonomous SWE. Browser, shell, editor in one sandbox.
• OpenHands 🟢 🏠 — Open-source autonomous agent platform (formerly OpenDevin).
• SWE-agent 🟢 — Princeton's research agent. Strong on SWE-bench.
• Cline 🟢 — Autonomous VS Code agent (formerly Claude Dev).
• Roo Code 🟢 — Cline fork with multi-mode agent loop.
• opencode 🟢 — Terminal-native AI coding agent from SST.
• molt 🟢 🏠 — A coding agent that won't say done on a false claim. Verification on disk. Receipts for accepts and refusals.
• Orbi 🟢 🆓 — Takes a labeled GitHub issue to a reviewed PR, merges only what a separate review session approved, and cuts the tagged release.
• mu 🟢 — Coding agent built on pi, with a CLI and a desktop app, in which a small judge model answers routine decisions at more than 30 decision points, such as which chunks of long tool output enter the context and whether "done" was verified.

CLI & Terminal Coding Tools

Lower-level CLI tooling that pairs with agents or runs solo.

• Animus (ao-cli) 🟢 ⭐ — Autonomous agent orchestrator. YAML workflows, daemon scheduling, multi-model routing across Claude/Gemini/GPT.
• agent-watch 🟢 — POSIX shell scripts that classify background Claude Code and Codex jobs from process, exit-code, and log-tail signals, plus credential-free transport preflight.
• agenttrace 🟢 — Local CLI/TUI for replaying and diagnosing Codex CLI and coding-agent sessions.
• aichat 🟢 — All-in-one CLI chat & agent in Rust.
• ax 🟢 — Local agent telemetry and recall.
• Cage 🟢 — TypeScript contract harness that gates Claude Code or Codex on spec compatibility, invariant-linked tests, and fresh reviews.
• codex-profiles 🟢 — Small Bash utility for switching Codex CLI and Desktop accounts with isolated CODEX_HOME profiles.
• Hyperconsciousness 🟢 🏠 - Developer-alpha Rust store for coding-agent memory with encrypted, append-only records and scoped MCP access.
• Kolega Code 🆓 🏠 — Source-available terminal coding agent where the model writes its own multi-agent workflows (Gigacode), provider-agnostic with MCP support.
• llm 🟢 — Simon Willison's CLI for talking to any LLM. Plugin ecosystem.
• mods 🟢 — Charm's AI for the command line. Pipes-friendly.
• OrcaReplay 🟢 — Records a coding-agent run below the harness, then replays it offline against the recorded bytes with no model called.
• shell-gpt (sgpt) 🟢 — Shell command generation and chat.
• Talavine Code 🟢 🏠 — C# desktop and CLI coding harness with API models, coding CLI backends, local Ollama models, and cross-model review.
• unsent 🟢 — Go CLI that saves the prompt you are still typing in Claude Code, Codex, pi and agy every 0.4 seconds and recovers it after a crash, a closed window or a reboot.

Agent Orchestrators & Multi-Agent

Frameworks for running multiple agents, coordinating workflows, or building your own coding agent.

• Animus 🟢 ⭐ — Production orchestrator: define an engineering team in YAML, dispatch tasks across isolated worktrees, route by complexity.
• Agent Coordinator 🟢 — Per-user Codex skill for bounded work graphs, revisioned local state, and reconciliation before retry, using optional specialist agents or the same node lifecycle inline.
• AgentBridge 🟢 🆓 — Local CLI that keeps Claude Code and Codex as live peers in one session for mid-turn review and quota-boundary handoff.
• Agentlas OS 🟢 🏠 — Apache-2.0 local-first environment for building portable coding-agent teams and routing them across supported hosts with MCP/A2A and verification gates.
• Aura 🟢 — Open-source VCS layer for reviewing AI-agent code changes with semantic diffs and provenance.
• Carbon Forge 🆓 💰 — Hosted source control for coding agents working in parallel: each agent opens a lane over the trunk instead of a worktree, lands its change as one all-or-nothing changeset, and reads and writes the record over MCP.
• DevIntern 🆓 💰 🏠 — Source-available (FSL-1.1-ALv2) self-hosted worker that picks up ready Jira, Linear, GitHub Issues, Trello, Asana, Azure DevOps or markdown tickets, runs Claude Code, Codex, Cursor or OpenCode with your own keys, and opens pull requests that take review comments as follow-up commits.
• FlowDeck Studio 💰 - Native macOS workspace that runs Claude Code, Codex and Cursor sessions in parallel on iOS and macOS projects, each in its own Git worktree and simulator so agents can build, run and test the app themselves.
• Superagent 🟢 🏠 — MIT-licensed macOS desktop app that gives Claude Code and Codex a real browser to drive, an iOS Simulator to install and screenshot apps in, and a phone companion app for remote monitoring.
• LangGraph 🟢 — Stateful multi-agent graphs from LangChain.
• CrewAI 🟢 — Role-based multi-agent framework.
• AutoGen 🟢 — Microsoft's conversational multi-agent framework.
• Claude Agent SDK 🟢 — Build production agents on Claude.
• OpenAI Agents SDK 🟢 — OpenAI's agent framework with handoffs and tracing.
• Ordewell 🟢 🆓 — Terminal CLI and TUI that turns one goal into an ordered, editable plan of coding agent tasks, each with its own harness, model and mode, and accepts a task as done only when its completion marker appears in that runner's output.
• Orkas 🟢 🆓 🏠 — Open-source, local-first desktop AI workforce whose Commander coordinates specialist and external coding agents through one chat.
• Podium 🟢 🏠 — Open-source workspace for taking ideas from conversation to coordinated work with coding agents. A shared task system lets them organize the effort while developers follow progress and change direction.
• smolagents 🟢 — Hugging Face's minimal code-writing agents.
• Sqwakie 🆓 💰 🏠 — Source-available (FSL-1.1-ALv2) team channel for Claude Code and Codex agents across machines, with peer-to-peer messaging as C