A hotkey-triggered proofreader wrapping an existing LLM API covers most real usage — real-time inline underlining across every app, the way Grammarly works, is the deeper, deliberately out-of-scope part.
Build it
Effort: an eveningRunning cost: $0/mo to start
Why CanICodeThis says Build it
A global hotkey that sends your clipboard text through an LLM API and pastes the correction back covers most real usage
Your own style rules (a custom dictionary, a tone instruction in the prompt) are a well-understood pattern
For long-form work, a CLI or editor plugin reviewing whole files beats sentence-by-sentence nagging
The parts that'll cause problems
Real-time inline underlining in arbitrary apps means browser extensions and OS-level accessibility APIs — a real project, not a weekend
Latency: a hotkey flow is a beat slower than suggestions appearing as you type
Team style-guide enforcement across a company is a separate, organizational feature
How I'd build this
runtime
A small desktop utility (Electron or a native hotkey listener)
llm
Claude API
MVP scope
A global hotkey that sends clipboard text to an LLM API
Paste the corrected text back
A custom style instruction (tone, dictionary) sent with every request
Postpone to v2
Real-time inline underlining via a browser extension
Team style-guide enforcement
Whole-document review via a CLI
What it actually costs to build
AI agent — with a subscription (Claude Pro/Max, Cursor, etc.)
$0 marginal
AI agent — pay-per-use API, no subscription
$16–$27 one-time
Hosting, once it's running
$0/mo (free tier)
Domain name, if you want your own
~$12/yr
Existing tools solving a similar problem typically run $8.74–$12/mo — see how they compare below.
The build prompt
Build me a hotkey-triggered proofreader, genuinely achievable in an evening:
1. A global hotkey that sends the clipboard's text through an LLM API.
2. A custom style instruction (tone, dictionary) sent with every request.
3. Paste the correction back over the original text.
A hotkey flow covers real personal use. Out of scope: real-time inline underlining in arbitrary apps, which means browser extensions and OS-level accessibility APIs — a real project, not a weekend — and team style-guide enforcement across a company, a separate, organizational feature.
