Private beta The VibeMyWay engine, now a native Mac app

Ship a week of content.
From one Mac app.

VibeMyWay Desktop runs the full agentic content pipeline natively on macOS. One brief in: blog post, email sequence, social package, landing pages, and images out. Review, rate, and ship from one place. No terminal, no cloud, no lock-in.

Request an invite

macOS on Apple Silicon · Bring your own AI provider · Invite only

VibeMyWay Desktop dashboard on macOS, showing three active campaigns, a drafts queue, indexed brand assets, agent run counts, a list of recent outputs with star ratings, and the Brand Copilot panel open on the right.
The full engineNo AI editor required
Bring your own AIClaude, ChatGPT, or any OpenAI-compatible API
Local-firstEvery file lives in a folder on your Mac
Keys stay yoursStored in the macOS Keychain, never written to disk
What it does

Your whole content operation, in four panes.

The workflow practitioners already run inside Claude and Cursor, now with a queue, a review pane, and a brand memory that persists between runs.

01 / Pipelines

One brief in. The whole pipeline out.

Answer once, then walk away. The Leads Pipeline turns a single offer into a lead magnet, a landing page, an email sequence, and a social package, generated in order, each step feeding the next.

  • Full pipeline or single step. Run all four stages in one pass, or jump to any step you already have the upstream output for.
  • Fast draft or full polished. A quick judgement pass with placeholder images, or every gate, methodology, and design pass up front.
  • Never re-asks. Answer once and every downstream step uses those choices.
The Leads Pipeline pane with Full pipeline selected, offering lead magnet formats, landing page type, email sequence length, and social atomisation options in a single form.
02 / Providers

Your models. Your keys. Your Keychain.

Plug in the AI you already pay for. A Claude or ChatGPT subscription, or any OpenAI-compatible endpoint. Keys go straight into the macOS Keychain, never written to disk and never shown again.

  • Subscription or API key. Anthropic, OpenAI, and any OpenAI-compatible provider you already use.
  • Keychain-secured. Credentials stay on your Mac, handled by the system, not by us.
  • You pay the provider directly. Their rates, their bill, with nothing added on top and no credits to buy from us.
The Providers and API keys settings pane, listing six configured providers including Anthropic and ChatGPT subscriptions alongside custom OpenAI-compatible models, with one marked primary.
03 / Model routing

The right model for every job.

Orchestration, writing, design, research. Each role gets the model it deserves, so a heavyweight handles campaign strategy while a faster model carries the volume work. Or hit one preset and move on.

  • Per-role routing. Assign any configured model to orchestration, writing, design, or research.
  • Balanced and Economy presets. Sensible tiers assigned across every role in one click.
  • Spread across models. Distribute the work when you would rather not lean on one provider.
The Model routing settings pane, assigning a model to each role: primary agent, orchestration, writing, design, and research, with Balanced, Economy, and Spread across models quick presets.
04 / Brand Copilot

A copilot that actually knows your brand.

The Brand Vault holds your voice, your logos, and your colours. The copilot reads all of it before it answers, so a brand voice check is an actual check against your rules rather than a polite guess.

  • Brand voice check. Audit any output against the voice you defined, before it goes out.
  • Rules that stick. What the kit learns gets applied before every generation, not re-explained each time.
  • Rated outputs. Star what worked. The runs you approve tell the next run what good looks like.
The Brand Copilot panel open beside the dashboard, offering brand voice check, content ideas, what to improve next, and sharper hooks as starting prompts, next to a list of star-rated recent outputs.
Review and ship

The part every other tool leaves to you.

Generating the draft is the easy half. The finished piece opens in a preview pane beside the run, so you read it, edit it, rate it, or send it back with a note in plain English. Nothing leaves your machine to get there.

A campaign view in dark mode showing 13 outputs, 19 minutes 15 seconds of run time and an estimated cost of 4.78 dollars, with the preview pane open on the right displaying a finished, published article with its hero image, a five-star rating, View and Edit tabs, a Revert to draft button, and a field to regenerate the piece with a note.

Read it as it will publish.

View and Edit sit on the same pane, with the hero image, headings, and body exactly as they will ship, so you judge the real thing rather than copying it into a second tool to find out.

Send it back in your own words.

"Headline weak." "Shorter." "Wrong tone." Type the note and regenerate. Tick Remember and that correction becomes a standing rule for everything after it.

Revert to draft, any time.

Published is a state you can undo, not a door that closes behind you. Rate what worked, unpublish what did not, and keep the version history on disk.

13Outputs
19m 15sRun time
≈$4.78Provider cost

One real campaign, measured by the app and captured in the screenshot above: 13 outputs across 3.4M tokens, billed by your own provider at their rates. This is an internal run on our own account, not a customer result or a guaranteed figure. Yours will differ with model choice and scope, which is exactly why the app puts the number on screen.

Local-first, by design

Your files. Your Mac. Your business.

VibeMyWay Desktop is a portable workspace. Everything it produces is a plain file in a folder you already own.

0
Cloud dependencies

Campaigns, outputs, and brand assets sit on your disk. There is no account to lose, nothing to sync, and no server of ours between you and your work.

100%
Portable outputs

Markdown, HTML, and images. Open them in any editor and publish them anywhere. There is no export step because nothing was ever locked up.

∞
Ownership

If you stop using the app tomorrow, every file you produced stays exactly where it is, readable without us.

The dashboard in dark mode showing a banner reading that a run was interrupted, with Resume and Keep what finished buttons, above campaign statistics and a list of recent outputs annotated with run times and costs.
Close the app mid-run and the work is still there. The run picks up where it stopped, because nothing was being held open in a session on somebody else's server.
Licence

One licence. No subscription.

The same deal as both kits. You buy it once and it stays yours.

What we can tell you now. VibeMyWay Desktop will be a one-time licence, exactly like ContentMind and MarketingKit. You buy it once, and it keeps working whether or not you ever buy anything from us again.

What we cannot tell you yet. The price. It is set at public release, and putting a number here during the beta would be a guess dressed up as a commitment.

What you pay for the AI. Whatever your own provider charges, directly to them. VibeMyWay never resells model access or marks it up, which is why the app shows you the cost of every run.

Request an invite

Fair questions.

What does the private beta include?
Access to the current build, the full set of pipelines, and a direct line to me when something breaks. New builds ship as they are ready. In exchange I want to hear what you hit: the beta exists to find the rough edges while the group is still small enough for me to fix them properly.
Do I need an AI subscription to use it?
You bring your own. Connect a Claude or ChatGPT subscription, or any OpenAI-compatible API key. The app never sits between you and the provider, so you pay them directly at their rates, with no credits to buy from us and no markup on top.
How is this different from ContentMind and MarketingKit?
The kits run inside Claude or Cursor, which means you need an AI editor and some comfort working in one. Desktop is a standalone app running the same practitioner-built engine, with pipelines, a review queue, brand memory, and model routing built in, so it runs on its own. If you already own a kit and like working in your editor, nothing changes for you.
Where does my content live?
In a folder on your Mac that you choose. Campaigns, drafts, brand assets, and images are plain Markdown, HTML, and image files sitting in normal directories. There is no database to export from and no account to close. Back it up the way you back up everything else.
What does a run actually cost?
Whatever your provider charges, and the app shows it. Tokens, run time, and estimated cost appear on every campaign, so the number is on screen rather than buried in a monthly bill. The campaign in the screenshot above produced 13 outputs across 3.4M tokens for roughly $4.78. That is one internal run on our own account, not a promise: your figure moves with the models you route to and how much you generate.
Is the app signed and notarized?
Not yet. During the private beta, builds go to invited testers and are not notarized by Apple, so macOS will warn you on first launch. Notarization lands with the public release. I would rather you heard that here than from Gatekeeper.
VibeMyWay Desktop for macOS

Your next campaign is one brief away.

The private beta is small on purpose. Tell me what you are trying to ship and I will send you a build.

No card, nothing to cancel. This is a beta list, not a newsletter.

  1. 1 You send this. It reaches my inbox, nowhere else.
  2. 2 I read it myself and check the fit against what the beta is testing.
  3. 3 You get a build or a straight answer, within two business days.

Your name, email, and message are used only to answer this request. See the privacy policy.

Prefer to work in your own editor? See the kits