RevampKeyRevampKey home
Download
AI CODE EDITOR

Build software with the discipline of a real team.Build software with the discipline of a real team.

RevampKey is an AI code editor that works like a product team. Describe what you want built: it researches, plans with you, designs the screens, writes the code step by step and tests it. Nothing gets built until you approve.

Download for macOSSee how it works

Linux too. Free to start.

Let clients book and reschedule their own appointments.
Researcherinvestigates
Product managerdecides
Designerdesigns
Engineerbuilds
QAchecks
Designer is drawing 3 screens for you to approve
In plain English

Think of it as a small product team that lives inside your computer.

YOU

Say what you want

Write it the way you would tell a colleague. No technical wording needed.

“Let clients book and reschedule their own appointments.”
THE TEAM

Five specialists get to work

One researches, one decides, one designs, one builds, one tests. When something is unclear, they ask you.

“How long should a time be held while a client confirms?”
YOU AGAIN

Approve before it is built

You see the plan, the screens and the decisions first. Nothing is built on a guess, and every step can be read back later.

“Approved: 3 screens. Build can start.”
Have an idea, no developer?Describe it, answer a few questions, approve the screens, and get working software.
Run a product or ops team?Get plans, reports and follow-ups you can read and challenge, not a pile of code.
Already write code?Use it as your AI coding IDE. Point it at an existing codebase: it maps the code, plans the change and commits in reviewable steps.
How it works

One idea in. A whole product team at work.

Follow one request through RevampKey’s AI coding agents, from idea to tested code: let clients book and reschedule their own appointments.

Read the full walkthrough of how the AI coding agents plan, build and test →

Researcher · investigates

Research lands as a report you can explore.

Before anything is planned, it reads your project and the web, then hands back a report. Not a wall of chat. Sort it, filter it, and see where every claim came from.

Reports / Booking readiness auditAUDIT · REV 1
RevampKey AUDIT

Booking readiness audit

What the appointments app handles today, and what self-serve booking needs first.

Four of the nine booking paths have no handler today.
9Paths in the flow
4Unhandled today
2Decisions for you
High

Two clients can confirm the same time at the same momentNo lock on a time slot today

High

A reschedule can leave the old booking behindThe old time is never released

Medium

A held time never frees if the browser closesHolds have no expiry

Medium

Reminders can go out at the wrong hour across time zonesTime is stored without a zone

Low

A cancel inside 24 hours is not flagged to the front deskNeeds your decision

Low

The confirmation has no calendar attachmentNice to have

More on interactive plans and reports →

Product manager · decides

It asks before it assumes.

Where your idea leaves a gap, it asks, one clear choice at a time, with a recommendation when it has one. Your answers are written into the plan, so nobody has to remember them later.

2 questions before I plan
How long should a time be held while a client confirms?
Can clients cancel inside 24 hours?
One more answer and it goes into the plan

Every path mapped. Every decision written.

The answers become a map of everything that can happen, including the unhappy paths, and a short list of rules the build must respect. You correct them here, while changing your mind is free.

Flow · booking lifecycle
Time pickedHeld for 5 min
Still free?Taken · offer next 3
Booking confirmedReminder sent
Spec · booking.md
Decisions
A time has at most one confirmed booking.
A hold lasts 5 minutes, then frees the time.
A reschedule moves the booking, never doubles it.
A cancel inside 24 hours is waiting for your answer.
Designer · designs

Screens you approve before any code.

You see what clients will see, including the awkward moments like “that time was just taken”, while changing it still costs nothing.

Pick a time
Pick a time

Shown in your time zone, London.

TUE 13 OCT
WED 14 OCT
09:3013:0016:30

We hold 11:30 for 5 minutes while you confirm.

Confirm booking
Just taken
That time was just booked

Someone confirmed 11:30 a moment ago.

11:30 on Tue 13 Oct is no longer free.

CLOSEST FREE TIMES
12:0015:00Wed 11:3011:30
Book 12:00
Booked
You’re booked

Haircut with Priya

Tue 13 Oct, 11:30

A reminder goes out on Mon 12 Oct.

RescheduleCancel
3 screens · version 2. Waiting for you.
Engineer · builds

Built in steps, each one checked.

The approved plan is split into small steps, each with its own checklist. Every finished step is saved as its own change you can read, so there is no giant pile of code at the end.

Plan · self-serve booking · 4 steps · 15 tasks
Slots and holds
4 tasks · 2 checks passed
a3f9c21feat(booking): hold a slot for 5 minutes
Done
Booking and rescheduling
5 tasks · 3 checks passed
7be04d8feat(booking): a reschedule moves, never doubles
Done
Reminders
3 tasks · 2 checks · working now
In progress
Front-desk view
3 tasks · 2 checks
Next
QA · checks

Checked by running it, not by reading it.

Each step is tested by actually using it, on every screen size, before the next one starts. A test that was never run is never reported as passing.

$ npm test -- booking
✓ a second client cannot confirm a held time
✓ an expired hold frees the time
✓ a reschedule leaves exactly one booking
✓ reminders follow the stored time
21 passed0 failed

Desktop, tablet and phone: every screen size rendered and looked at.

You answered 2 questions and approved 3 screens. The plan, the decisions and every change are saved in your project, so anyone can see why it was built this way.

Where it all lands

Every line of code has a reason. And a test that proves it.

By the time RevampKey’s AI coding agent writes code, it already has the research, every path through the feature, the decisions you made, the screens you approved and your team’s conventions. The code is built from all of it, then tested against the same rules, one module at a time.

ReportKnows what is broken

The research found holds that never expire, so every hold in the code gets an end time.

FlowHandles every path

“Taken” and “expired” were mapped before coding, so they are built now, not found later.

SpecEnforces your decisions

“One booking per time” becomes a check in the code and a test that guards it.

ScreensMatches what people see

When a time is taken, the code leads to the “just booked” screen you approved.

SkillsFollows your conventions

Your playbook says store times in UTC, so the code does, every time.

ModulesLands in checked steps

Each module has its own criteria and tests, and is committed once they pass.

Schedules

Tell it when. It comes back on its own.

Ask your AI coding assistant for a follow-up in plain words, or use cron if you prefer. It runs in the same project, with everything it already knows, and posts what it finds.

  • Plain words or cron. “Check the deploy in 5 minutes” and “every weekday at 9” both work.
  • Right time all year. Schedules stay correct when the clocks change.
  • Honest about what survives. “Always” schedules survive a restart. “This window” follow-ups end when you close it, and say so.
  • It sets its own follow-ups too. While it waits on something slow, like a deploy, it schedules a check and comes back. Those are marked Agent.
  • Nothing silently missed. If the app was closed, the run happens once on launch and is marked late.

How to schedule AI coding tasks in plain English or cron →

New schedule
Every weekday at 9, check yesterday’s bookings and flag anything odd
WhenWeekdays · 09:00
Next runNext weekday, 09:00
Runs inBookings app
Check overnight bookings and flag anything oddWeekdays · 09:00 · set by youAlways
Check the deploy againIn 5 minutes · set by the agent while it waitsThis window
Draft the weekly summaryMondays 08:30 · app was closed · last ran 10:12, lateRan late
Skills

Teach it how your team does things.

A skill is a short playbook in a plain file: how you write, test, review or release. Put it in your project once, and the AI coding agent follows it whenever the work calls for it, so you stop repeating yourself in every chat.

  • Loads when it is relevant. The agent reads each skill’s one-line description and opens the full playbook only when the task matches.
  • Plain files in your project. Keep them in .revampkey/skills or .agents/skills, and review and version them like code.
  • Works with skills you already have. It uses the common SKILL.md format, a name and a description at the top, so a skill written for another coding agent works once it is in your skills folder.
  • The whole team follows it. When work is handed to a specialist, the skills it needs go with it.
  • You can see it happen. The chat shows each skill as it loads.

How skills work and what a SKILL.md looks like →

All connected

Plans, schedules and skills work as one.

Your playbook shapes the plan. The plan drives the build. What comes back is a report. And a schedule sends the team back to look again, without you asking.

Anything new it finds becomes a question for you, and the plan updates. Then the loop continues.
MON 09:00The schedule wakes up in the Bookings project.09:00It loads your “Weekly metrics” skill, so it already knows where the numbers live.09:02It checks last week’s bookings and finds two reschedules that left duplicates.09:04It writes a report you can open, sort and filter.09:05It asks you one question, adds the answer to the plan, and schedules the next check.
Right-sized

A small fix stays small. A big change gets a plan.

There is no mode to switch on. RevampKey decides how much structure each request needs.

Small

Fix a typo, rename a button

Done and checked in a minute. No ceremony.

  1. Do it
  2. Check it
Medium

Add a feature to one screen

A short plan, the screens if it is visual, then built and tested.

  1. Ask
  2. Plan
  3. Screens
  4. Build
  5. Test
Large

Take payments, send reminders

The full team: research, every path mapped, approved screens, staged build.

  1. Research
  2. Flow
  3. Spec
  4. Screens
  5. Build
  6. Test
The real app

This is what you will see on your desktop.

The RevampKey AI code editor on one screen: chat on the left, reports, flows, screens and code in the middle, and the plan’s modules and schedules on the right.

For developers

Under the hood, for people who read diffs.

A desktop AI coding IDE built on Code - OSS, so your editor habits carry over. The AI coding agents work in your own repository, on your computer.

Works in existing codebases

Reads multi-repo workspaces and keeps a map of your classes and functions, so changes fit what is already there.

How it reads your code →

One commit per step

Each finished module is its own reviewable commit, tied to the checks it had to pass.

How changes land →

Specialists, isolated

A researcher, a coder, a designer, a Git operator and a tool connector, each with its own context.

How delegation works →

Plans live in your repo

Specs, trackers and diagrams are files in your project with revision history. Nothing is stored on our servers.

Where plans are kept →
Privacy

Your work stays on your computer.

Local first

Plans, specs and diagrams are written into your own project folder, not our cloud.

Never used for training

Nothing you work on is used to train models.

Keys stay in your keychain

Tokens for connected tools are kept in your operating system’s secure storage.

Read the privacy policy

Questions

Before you download.

Do I need to know how to code?

No. You describe what you want in plain words and review plans and screens, which need no coding knowledge. The code is written, tested and saved for you. Some familiarity helps on larger projects, and anyone who reads code can open every change it made.

What is an AI code editor?

A code editor with AI built in that can read your project, plan changes and write the code with you. RevampKey is a desktop AI code editor whose AI coding agents also research, map the flow, design screens and test the work before it is committed.

Is RevampKey an AI coding agent or an AI coding assistant?

Both, inside one editor. You talk to one AI coding assistant, and it hands work to specialist agents (research, design, engineering, QA) and brings the results back to you to approve.

Does it work with my existing codebase?

Yes. Open your repository and it maps the code before planning a change. Large changes are planned and built in small, reviewable commits. How it reads your code

What can I build with it?

Websites, apps, internal tools and automations, and changes to software you already have. Small requests stay small; bigger ones get a full plan.

Is my work private?

Yes. Your plans and files live in your own project folder, nothing is used for training, and secrets stay in your computer’s keychain. Read the privacy policy

What does it cost?

Every account starts with limited free usage and no card required. See pricing for plans and add-on credits. See pricing

Start

Start with one idea.

Describe something you have been putting off. Read the plan, approve the screens, and decide whether to let it build.

Download for macOSWindows (.exe)Linux (.deb / .rpm / .tar.gz)

macOS 12+ · Windows 10+ · Ubuntu 20.04+  ·  See what shipped in 2.8.0