web2mac
Turn any local web dashboard or tool into a macOS app that starts its own server, shows it in a native window, and stops it when you quit.
$1
Works with the AI Agents you already use
web2mac
Example session with this skill installed
Turn the static dashboard in ./static-dashboard into a Mac app called "Static Dashboard" that I can keep in the Dock.
- Read your context and instructions
- Compiled the web2mac
Static Dashboard.app is ready
The folder only has an index.html, so detect.py chose
static mode: the app serves it itself. I wrote a config, built the app and checked its lifecycle (this is a real run, not a mock-up).
web2mac.json
{
"name": "Static Dashboard",
"mode": "static",
"static_dir": ".",
"port": 0,
"icon": { "emoji": "📊", "top": "#3A6FF7", "bottom": "#1B2E7A" }
}
Build
built : ~/Applications/Static Dashboard.app
mode : static port: 0
run : exec /usr/bin/python3 -m http.server {port} --bind 127.0.0.1
log : ~/Library/Logs/StaticDashboard.log
Lifecycle checks
| Check | Result |
|---|---|
Launch: log shows started server (pid …) on port … and page loaded in the window | ok, http 200 |
| Quit with ⌘Q | server stopped, port free |
kill -9 on the app | server gone within ~2 s (watchdog) |
| A server already running on a fixed port | log says adopted; it keeps running after quit |
| A command that does not exist | dialog with exit code 127 and a button for the log |
Next steps
- Drag
~/Applications/Static Dashboard.appto the Dock. - Port
0picks a free port on every launch, so a stale server is never reused. - For a server-based tool (Streamlit, Vite, Flask...) use
mode: commandwith{port}in the command; the skill proves the command answers before building.
Connects securely to your tools. The creator never sees your data.
About this skill
web2mac wraps a local web tool (Streamlit, Vite/Next, Flask/FastAPI, a plain index.html, anything that serves HTTP on localhost) in a real macOS app with its own Dock icon and window.
The app manages the server for you:
- Starts it with the command you configure, through a login shell so pyenv/nvm/Homebrew are on the PATH. If a server is already running on the port, it adopts it and never kills it.
- Waits until the server answers, with a splash screen, and shows a clear dialog with the exit code and the log if it fails.
- Stops it when you quit. If the app is force-quit or crashes, a small watchdog still kills the server, so nothing is left holding your port.
No Electron and no npm: the generated app is a small Swift binary hosting a WKWebView.
What the skill does when you ask Claude to "turn this dashboard into a Mac app":
- Detects how the project starts (
detect.py). - Writes a
web2mac.jsonconfig. - Runs the command on a spare port and checks it answers, before building anything.
- Builds the
.appwith an icon (emoji, initials or your own image). - Verifies the lifecycle: quit frees the port,
kill -9still cleans up, an existing server is adopted.
The window also gives you: working copy/paste/undo, reload, back/forward, zoom, file downloads to ~/Downloads, file pickers, native alert/confirm, external links opening in your browser, and remembered window size.
Requirements: macOS 13+ and the Xcode command line tools. The machine needs the runtime your command uses (Python, Node...); nothing is bundled. The app is ad-hoc signed, fine on the Mac that built it; sharing it with others needs Developer ID signing and notarization, which this skill does not do.
Open source (MIT). The full source is free at https://github.com/igerm/web2mac. The price here is a "thank you" tip, not a paywall.
Requires macOS 13+, Xcode command line tools. Does not support subfolder scanning during detection. Requires a login shell for command execution.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 11 days ago
Requires