HewyDownload
Pixel art: Hewy's green wren standing on a small home server on an oak desk by a sunny window, beside a mug, a notebook and a fern.

Your AI agents, on your own servers.

Claude, Codex and OpenCode on your servers: a real app to steer them, a real terminal when you want one, a browser they can use when you let them, and approvals from your phone.

Get it forAndroidMacComing soon foriPhone

Hewy on a Mac: servers in the sidebar with their chats, each chat with its own pixel wren, a few at work and one waiting under Needs you, and a chat open on the right speeding up an orders query. An agent on web-01 asking on an iPhone to run the checkout tests: Yes; Yes, and don't ask again for npm test; or No.

AI agents

Close your laptop. Your agents keep working.

Claude, Codex and OpenCode run on any server you can SSH into, right next to your code, and keep going while your laptop's closed and your phone's asleep. Talk to them, or just say what you want with the microphone.

  • Use your Claude or ChatGPT plan, or your own API keys.
  • Pick the model and effort for each chat, and switch any time.
  • Give it a goal, and it works until it's done.
  • Repeat a prompt every night, or fork a chat to try another way.
  • See how much of each Claude or ChatGPT plan is left, account by account.
  • Stop always stops. A stuck agent is restarted, and the chat carries on.
A chat with Claude on api-01: it added rate limiting to the public API and checked it. The message box is taking dictation, its waveform moving, and below it the chat runs Opus 5.5 at high effort on a Claude plan. The model menu: Claude with Opus 5.5 chosen, Sonnet 5.5 and Haiku 5.5; Codex with GPT-5.5, GPT-5.5 Codex and GPT-5.4 mini; and OpenCode with models from Anthropic, OpenAI and Google.

Web control

Stop screenshotting for your agent.

The browser beside each chat lets your agent look for itself: your dev site, your dashboard, the replies to your post, the store console you're stuck on. It clicks through like a user and tells you what it found, in your own browser, where you're already signed in.

  • Checks its own work on your dev site, on any server.
  • Tests sign-up, checkout and settings flows for you.
  • Reads replies, dashboards and consoles. No copy and paste.
  • You choose per site: Off, Can look or Can drive. Stop is always there.
  • Asks before it buys, sends, posts or deletes. Never sees your passwords.
  • Claude, Codex and OpenCode. Mac and Android today, iPhone soon.
A chat on web-01 asked to check that checkout still works after the cart change. Beside it, the shop at localhost:3000 on web-01, with the agent's green cursor and wren clicking Add to cart, and a bar at the bottom: Merlin is clicking Add to cart, with Stop. A chat asks to use status.example, to check whether last night's API errors show on the status page, with a reminder to sign in first if you need to: Don't Allow or Allow.

Any agent

When one agent won't do the work, another will.

Every AI provider has its own rules, and they change. A security review of your own product, a penetration test, an ordinary fix that trips a filter: it worked yesterday, and today the agent stops. Switch the chat to Claude, Codex or OpenCode, and the conversation so far goes with it, so the next agent picks up where the last one stopped.

  • Switch agents in the middle of a chat.
  • OpenCode reaches models from Anthropic, OpenAI, Google and others.
  • No one provider's rules decide what gets done.
  • Your code stays on your servers, whichever agent you use.
  • One app, one usage view, for every provider.

Connections

Agents that work together, with you in charge.

Connect a chat on one server to a chat on another, and their agents can message each other: your Mac's app agent asks your web server's agent for an API change, and hears back when it's live. No chat can reach another until you connect them.

  • You choose who can message whom, and which way.
  • Messages only, or images and files too.
  • For an hour, a day, or until you remove it.
  • Carried by your own devices, never our servers.
A chat on api-01 gets a message from Bizness Bob on web-01: The Pro plan launches Monday. Can the public API give Pro keys 600 requests a minute, and keep free keys at 120? The agent makes the change, the tests pass, and it sends word back, delivered. The connection between Bizness Bob on web-01 and Ragnar the Rate Limiter on api-01: both ways, messages and images, with the three messages it carried, each delivered.

Beside the chat

Terminals, files and web pages, right beside the chat.

When an agent gives you a command, one tap opens a terminal on that server with it typed in, ready for you to read and press Return. Paths open in Files, reports open as documents, and links open as pages. On a Mac, show the chat alone, the two side by side, or the pane alone. The same tabs follow the chat to your other devices.

  • Run in Terminal on any command, on this server or the one it names. Never run for you.
  • Markdown, code, PDFs and images, open beside the chat.
  • QR codes in replies, ready to scan with your phone.
  • The same tabs on every device. Sign-ins stay on each.
Hewy on a Mac: a chat on web-01 that deployed the shop's dashboard, and beside it three tabs, the Shop admin page at localhost:3000, the deploy's terminal and README.md. The same three tabs sit under the chat in the sidebar. The same chat on an iPhone, its tabs in a menu: localhost:3000, deploy@web-01: ~/shop and README.md, and below them, open a new browser tab, terminal or Files.

Terminal and files

Terminals and files, on every server.

Shells that keep running after you close the app, and every server's files to browse, preview, upload and download, on your Mac, iPhone and Android.

  • Snippets: the commands you use often, a click away in any terminal.
A terminal on web-01 with its snippets above it, Restart shop, Tail logs and Deploy, and below, the shop's build and its tests passing, all 42 of them.
Files side by side: a project on this Mac and /srv/shop on web-01, with a release uploading from one to the other.

VPN and port forwarding

Go online through any server, or bring its ports to you.

Send your Mac or iPhone through a server in one tap, as if you were on its network, or open a server's dev site, database or dashboard right on your device. It all runs over the SSH you already have, with nothing to install.

  • All your traffic, or just the server's network.
  • Forwards you switch on from the sidebar.
  • Your traffic goes only to your own server.
hub's VPN on All: connected, 1.35 GB through 23 connections.

Privacy

Private by design.

Your devices connect straight to your servers, never through ours. What syncs is encrypted on your device first, and your SSH keys never leave its security chip. When an agent uses your browser, it reads only what's on the page for you to see, never your passwords.

How Hewy protects you →

Make it yours

A little helper in every chat.

Every chat gets a pixel wren in its server's color, and a look all its own. Rather have plain icons? Turn them off in Settings.

Pixel art: a small home server on a shelf with Hewy's green wren on top, while someone reads in an armchair nearby.

Your agents, on your servers, never clocking out.

Free for 14 days, then US$99 a year or US$9.99 a month.

Get it forAndroidMacComing soon foriPhone

Download Hewy

Questions

Do I need to keep my computer on?

No. Agents run on your server, which is on anyway. Your devices only connect to check in, approve and pick up where you left off.

Which servers work?

Anything you can SSH into. Agents run on Linux servers and Macs. Routers, NAS boxes and other devices that speak SSH get terminals, files and key management.

Which AI can I use?

Claude, Codex and OpenCode, with your own API keys or your Claude or ChatGPT plan. You choose per chat, and can switch a chat to another agent or model at any time.

Why not just use Claude's or ChatGPT's own apps?

Keep using them; Hewy works with the same plans. What it adds: the agents run on your own servers, next to your code, rather than in someone else's cloud; one chat can move between Claude, Codex and OpenCode when one won't do the work; approvals reach your iPhone, signed with its key; agents on different servers can work together; any of them can use a browser on your own devices, your phone included, with the same rules for every agent; and one view shows what every provider has used. Plus a full SSH client for everything else.

Can my agent use websites?

Yes. It gets a browser tab beside its chat, on your own device, where you're signed in, so it can check its own work on your dev site, click through a flow as a user would, or read a dashboard or the replies to a post, without you pasting screenshots back and forth. You choose for each site whether it can look or drive, Hewy asks before it buys, pays, sends, posts or deletes, the agent never sees or types your passwords, and you can stop it any time. It works with Claude, Codex and OpenCode, in Hewy for Mac and Android today, with iPhone coming in an update. On a Mac it keeps working in the background; on Android, while Hewy is open. How it's kept safe →

What if an agent refuses to do something?

Switch the chat to another agent. Each provider has its own rules about what its models will do, and they don't always agree, so work one turns down another will often take on. The conversation so far goes with the switch (the most recent part, up to about 24,000 characters, with the tools that were used), so you don't start over.

Can Hewy see my code or my conversations?

No. Your devices talk straight to your servers. What syncs between your devices (your server list, saved passwords and API keys) is end-to-end encrypted, so our servers can't read it. What an agent reads goes to the AI provider you chose, as with any coding agent.

How much does it cost?

US$99 a year, or US$9.99 a month, after a 14-day free trial, for every feature on all your devices. AI usage is billed by your AI provider; Hewy never marks up tokens. Pricing →

Does it run on Android? And Windows?

Android, yes: download Hewy for Android here on hewy.ai. The Mac app downloads here too, and iPhone is coming to the App Store. Each is an app of its own, and your Hewy account keeps your servers, keys and settings in step between them, end-to-end encrypted. Windows as a server, yes too: Hewy sets up and controls agents on Windows 10 and 11 PCs, alongside Linux and Mac servers. There's no Hewy app for Windows itself for now.