Intent Keyboard

January 28, 2026

You have an idea. You want it done. But first: open the app, navigate the menu, configure the thing, wait for it to run.

What if you could just press a key?

"I'll run tests and deploy preview"
VS Code → auth.tsx → feature/auth branch
Running 12 tests
Building for production
Deploying to preview
Posted in #dev

Try it

Press spacebar. Watch the command execute. See the text change.

press space

The intent keyboard transforms how we interact with software. Press one key, and an AI agent figures out what you need.

One key

Multiple keys mapping to different agents breaks down. Intent is fluid. Context is heavy. You end up maintaining a mental model of which key does what in which situation.

One key is simpler. Press it, and an AI agent gets full context: screen state, active app, recent history, project state. The agent decides what action makes sense right now.

Press = "Take over and move me forward"

The demo

You just committed auth changes. Press the key.

"I'll run tests and deploy preview"
Running 12 tests
Building for production
Deploying to preview
Posted in #dev
Done in 4.2spreview-auth-7f3k.vercel.app

If tests fail: creates ticket, assigns you, links to logs.

One key with verbs

The key doesn't represent an app. It represents a contract: "Take over and move me forward." Different gestures give you different verbs.

TapDo the next obvious action

Default best guess based on context

HoldExplain / summarize / propose

No execution, just show options

Double-tapRepeat last successful intent

Same action, fresh context

Shift + tapLocal only

No network, no external APIs

Option + tapDry run

Show plan, require confirmation

Cmd + tapCompose

Draft output: email, PR, ticket, message

Agency ladder

Trust is what kills agent buttons. You need to know what it can do and when it asks permission.

1
Suggest

Shows what it would do. No changes.

2
Draft

Creates output. You approve before send.

3
Execute reversible

Runs actions that can be undone.

4
Execute irreversible

Always confirms first. Deletes, payments, external posts.

5
Autopilot

Timeboxed, cancellable. You set the window.

Receipts

Every action has a receipt. You can see what it saw, what it decided, what it changed, and how to undo it.

Receipt #4821
Context
VS Code → auth.tsx → just committed → branch feature/auth
Decision
Deploy to preview (confidence: high)
Actions
npm test → 12 passed
npm run build → success
vercel deploy → preview-auth-7f3k
slack post → #dev
Rollback
vercel rollback preview-auth-7f3k

Same key, different context

The agent adapts. Same key, different situation, different action.

VS Code, just committed
Runs tests → builds → deploys preview → posts in Slack
Figma, components selected
Exports SVG → updates design system → notifies team
Email open, meeting mentioned
Extracts details → checks calendars → sends invite
Doc finished, publish intent
Formats → generates OG image → schedules posts

Not macros

Macros execute fixed sequences. This reasons about context and chooses the right action.

Macro
if (keypress) {
  run("deploy.sh")
}
Fixed. Breaks when context changes.
AI Key
context → reasoning → action

Adapts to what you're
actually doing
Adaptive. Context-aware.

The bet

The keyboard is a wedge. The real product is the context layer (what's on screen, in repo, in tools), the policy layer (constraints, approvals, audit), and the workflow runtime (tools, connectors, retries, state).

If pressing the key reliably does something high-leverage with low latency and high trust, the concept isn't pointless. Software becomes something you command, not something you operate.