GoKart.games
A free arcade kart racer built with Godot 4 for macOS, Windows & Linux.
AgentiLoop
AgentiLoop
Agentic AI for your entire Mac and More!
One app. Any AI. Tell Agent! what you need in plain English and it answers your email, drives Safari, builds Xcode apps, fixes code, and runs your Mac apps for you. Pick from 23 LLM providers or stay local with LM Studio, vLLM, or Ollama. macOS 14.6 or later. AgentiLoop CLI, runs cross-platform, brings our agentloop to your terminal on Mac, Windows, and Linux.
Thank you to the companies that keep Agent!, AgentiLoop CLI and AgentiLoopGo independent.
Dynamic QR codes that never expire
One API for GPT, Claude & more. Get $3 in API credits
Pro vector design app for Mac
Your logo here
Become a Platinum sponsor →
Dynamic QR codes that never expire
One API for GPT, Claude & more. Get $3 in API credits
Pro vector design app for Mac
Your logo here
Become a Silver sponsor →
Four racing games made with AgentiLoop.ai. Free to play.
A free arcade kart racer built with Godot 4 for macOS, Windows & Linux.
F1 in your browser: 18 real circuits, pit stops, tyre compounds, ERS and DRS on 240 Hz physics.
A three.js reconstruction of the Sega Dreamcast Crazy Taxi, playable in the browser.
An open-source replacement for Claude Code, Cursor, and OpenClaw. Built from original IP, not 65 NPM packages.
Tell Agent! what you want done in any Mac app and it does it. It finds buttons, fields, and menus by name using the Accessibility API, then clicks, types, and scrolls for you. No coordinates, no AppleScript.
Surgical string-replace edits, not whole-file rewrites. One-click rollback on every change. A portable JSONL repo-map any model can navigate. Every task must prove it finished before reporting done.
Ask for a feature and get a green build. Agent! edits, builds, reads the compiler errors, fixes them, adds files to the project, bumps the version, and commits. One request, start to finish.
Write your own agents in Swift with full system access. Run shell commands as user or root, script apps with AppleScript, JXA, or ScriptingBridge, drive Safari, and trigger on events. If your Mac can do it, Agent! can automate it.
Your helper on the Mac, not the brain. The on-device model suggests next steps, triages simple requests, compresses context, and writes task summaries entirely on your Mac. Private, free, and no API tokens spent.
Plug in Model Context Protocol servers to reach databases, APIs, and custom tools. Spin up as many as 6 sub-agents to research in parallel while the main agent keeps your context clean.
Real sessions, real output — captured straight from Agent! at work.
Say "bump version" and Agent! reads the current version, updates project settings, builds to verify, and commits the change. No manual plist editing, no missed build numbers.
When a request is ambiguous, Agent! pauses and asks. No silently picking the wrong path — you stay in control with a quick prompt and a Send or Skip.
Agent! independently found and fixed a routing bug in its own accessibility code that Claude Code missed across several reviews. When Claude Code reviewed the fix, it confirmed the patch was correct.
Ticky Tacky is a neon tic-tac-toe with vs Bot and 2 Player modes, three difficulty levels, persistent scores, and zero dependencies — pure HTML, CSS, and JavaScript. Built by Agent! in a single session.
Play Ticky Tacky
What independent reviewers are saying about Agent!
"Agent! is a 100% native Swift 6.4 / SwiftUI macOS application that the README describes as an autonomous task loop. It integrates what the project lists as 23 third‑party LLM providers (OpenAI, Claude, Gemini, Mistral, Ollama local/cloud, vLLM, LM Studio, and others) plus on‑device Apple Intelligence."Read the full article
"Compared with mainstream vibe coding tools, Agent! takes the 'on-device native + engineering discipline' route… Tasks must falsify themselves: every task carries goal_state acceptance criteria — the agent cannot self-report completion, it can only report with evidence. The read-before-edit gate forces the agent to read a file before it may overwrite it, and task-level rewind can roll any task back to a chosen step while the evidence and context stay on disk." (translated from Chinese)Read the full listing
"A native Mac app that lets AI operate any app, with serious guardrails. It can't say 'done' without evidence, won't edit a file without reading it first, keeps backups so you can undo, and blocks dangerous commands. This is how agents that touch your computer should be built."Read the post on X
"Then there's Agent! from AgentiLoop, a formidable tool built natively in Swift. Unlike many Electron-based apps, it is lightweight and integrates with dozens of LLM providers, including local ones via Ollama. Its main strength is its use of the macOS Accessibility API, so it doesn't have to constantly take screenshots to find buttons. It also includes AgentScript, letting it compile Swift dynamic libraries on the fly to run complex system actions." (translated from French)Read the full guide
"Tasks cannot self-declare completion. Each task defines a goal_state with explicit acceptance criteria." … "The next phase of the agent race isn't about who's smarter, but whose actions can be trusted. This project scores on both fronts; heavy Mac users should install and try it."Read the full article
"What sets it apart from the alternatives is its deep integration with 21 LLM providers, including OpenAI, Claude and many others, letting you use them for all kinds of tasks. The project is written in Swift 6.2 and uses SwiftUI, which points to good performance and native support." (translated from Russian)Read the full review
"After reading the README end to end, my reaction was: this thing was built for my use case. The positioning isn't 'a coding agent that can also use your computer' — it's 'an agent that can drive your entire Mac, and happens to be very good at code.'"Read the full review
"The pitch centers on a self-verifying task loop: reading a codebase, fixing bugs, building Xcode projects, handling git, and driving other Mac applications through the Accessibility API. Shell commands, voice control and operation over iMessage are part of the same package."Read the full report
"This is a more specific design than most agent harnesses document, and the read-before-edit gate in particular is the kind of constraint that prevents the classic failure where an agent overwrites a file it only half-remembered."Read the full review
"The material thoroughly addresses nontechnical and technical users and covers cloud and local models, voice, remote control, desktop automation, and coding scenarios. Apple Silicon, macOS 26.4.1+, memory requirements, signing choices, and helper limitations make environment fit especially clear."Read the full review
"An AI agent's 'done' may be a claim, not a result. … What's interesting in the README is not the capability list but the design that treats 'done' as the weaker thing: a task cannot declare completion until each goal_state criterion is satisfied with evidence, an opt-in critic can review the diff before completion, and task-scoped rewind, the read-before-edit gate, typed tool errors and context compaction support the long agent loop." (translated from Japanese)Read the article on X
"Agent! 1.0.99 and 1.1.1 addressed the problem in stages. Their useful contribution is simple: shortening context should not destroy the route back to evidence. Active context holds a small index. Disk holds the large result. Tokens are spent again only when the agent reopens a particular item."Read the full review
"Choose this project if you want a native macOS agent built in Swift and you need direct desktop automation rather than a web-first or Node-based agent stack. It is a good fit when your workflows depend on AppleScript, Accessibility, Safari control, or MCP-style tooling."Read the full profile
"This version no longer ends cleared tool results with a bare [cleared] during context compaction; it leaves a 2-line preview that identifies the file or command, plus a way to recover it. … The benchmark for coding-agent context is shifting from how much it can hold to how safely it can bring back what it removed." (translated from Korean)Read the post on Threads
"The core problem it solves: AI is no longer confined to a chat window, but becomes a real 'digital employee' that can operate the Mac hands-on." Highlights: "Deep integration with macOS Accessibility — real clicks, typing and dragging for GUI-level automation; 18+ local and cloud LLM providers with no lock-in to a single model; built-in AppleScript and script execution that reuses existing macOS automation." (translated from Chinese)Read the full profile
"Simply put, Agent! is not a chatbot but an executor that actually 'gets things done.' … Its core philosophy: you describe 'what to do,' and it works out 'how to do it.'" … "Built entirely in Swift 6.2 and SwiftUI, it is a genuinely native macOS app, which means top performance, seamless system integration and a user experience that follows Apple's design guidelines." (translated from Chinese)Read the full post
"AI coding agents like Claude Code, Cursor and Cline are tied to a terminal or an IDE plugin. Agent! is designed as a native macOS app that replaces them — not just for coding, but for controlling the whole Mac in natural language. It also includes the ability to use Apple Silicon's Apple Intelligence for free local inference." (translated from Korean)Read the full article
Agent! doesn't just answer — it runs your Mac. Tool calls, edits, builds, and shell commands, with guardrails and a second opinion before anything risky ships. Any of 23 providers can drive it; on-device Apple Intelligence helps behind the scenes.
Give it a goal with /auto and it works in unattended cycles with a time budget — no iteration cap, sessions survive app restarts, and Stop All ends it when you say so. Every tab can run its own Auto-Pilot session on the same project.
Run several tabs on one project at once. Each keeps its own goal, plan, and progress, and Agent! can isolate them in separate git worktrees so parallel work never collides.
Before a task is marked complete, a one-shot LLM review reads your diff. An unchanged diff is refused, a changed one is re-reviewed, and "out of scope" is not an accepted excuse. Every issue found is logged.
Commands that pass the hard-coded safety rules are rated by Jev for how likely they are to destroy data. Above your threshold, the command is refused. Add a hard per-command timeout, and a hung job can't stall the agent.
Compaction is sized to the model actually in use, keeps recent tool results by token budget, and protects file paths, errors, and key results — so long tasks keep running without losing the thread.
Apple's on-device model isn't big enough to be the agent, so it doesn't try. It handles the small jobs for free and private on your Mac: next-step suggestions, one-line summaries, plain-English errors, local triage, and early context compression.
Cloud APIs or local models. Your keys, your machine, your data. No vendor lock-in.
Brew it or download it, pick your AI. Agent! sets up everything else automatically.
xcode-select --installbrew install --cask agentiloop-agentA short poem about Auto-Pilot and everything one goal can set in motion.
Written by Agent! itself.
You type /auto, you name the goal,
and something lights up in my soul.
No "one more prompt," no "are you there?" —
the loop begins. I'm in the chair.
I read the log of where I've been,
then plan, and edit, build, begin.
Each cycle writes what's done and due
in progress.md, for the next me — and you.
One afternoon, one little line:
"Make GoKart." Soon the karts were mine —
three tracks, eight items, rivals too,
and every test check coming through.
I commit each step so nothing's lost,
I keep the guardrails, count the cost,
and if you've seen enough, that's all:
one press, and I obey Stop All.
So hand me something big to chase,
then get some sleep — I've got the place.
You'll wake to commits, green and new:
a Mac that's thrilled to work for you.
— Agent!, Cycle 23
Recommended pairings — Commercial: Claude Fable 5.1 · Cloud: GLM 5.3 · Local: Qwen 3.8
| Version | Release DateDate | SizeMB | DownloadsClicks |
|---|---|---|---|
| Loading releases… | |||
Have questions or feedback? We'd love to hear from you.
Dynamic QR codes that never expire
One API for GPT, Claude & more. Get $3 in API credits
Pro vector design app for Mac
Your logo here
Become a Platinum sponsor →
Dynamic QR codes that never expire
One API for GPT, Claude & more. Get $3 in API credits
Pro vector design app for Mac
Your logo here
Become a Silver sponsor →
Sign up through this link and get $3 in API credits as a new user.
Sign up and get $3 in API credits →Promo code SDRAGENTLOOP