SnipMD turns any screen content into editable Markdown with on-device OCR — and transcribes audio & video locally with Whisper. Cloud AI translation, math-to-LaTeX, subtitles, smart redaction, and 11 UI languages. No API keys to configure.
See it in action
Watch a Stack Overflow answer go from Screenshot 2026-07-18.png to well-formatted Markdown with code blocks preserved — no typing, no re-formatting.
Features
On-device OCR and Whisper transcription, cloud AI tools, subtitles, batch automation — all from one menu bar icon.
⌥⇧3 ⌥⇧4 ⌥⇧5 capture, ⌥⇧V paste OCR, ⌥⇧P math, ⌥⇧I import media — chosen to avoid the macOS ⌘⇧ system shortcuts.
25+ languages, fully offline. Apple Vision under the hood, with smart re-OCR (grayscale + 2× scaling) on low-confidence reads.
Drop in an mp3 / m4a / wav / video file — Whisper runs entirely on-device (Apple Core ML) and returns a timestamped transcript in seconds. Clips up to 30s are free.
Every transcript segment is a [mm:ss] link. Click to jump and play; the active row highlights as the media plays. Videos also get on-screen text OCR'd every 5 seconds.
Export millisecond-accurate SRT for free. Pro adds one-call bilingual SRT translation and subtitle burn-in straight into the mp4 — no FFmpeg, fully sandbox-safe.
A 200-line heuristic engine turns OCR output into well-formatted Markdown: headings, lists, code blocks, paragraph breaks. No more walls of plain text.
Streamed, domain-aware translation that auto-detects medical / legal / software / financial / academic terminology — so 线程 stays 线程, not 运程线. Included with Pro, nothing to configure.
Snap a region with an equation. A vision model reads it and returns clean LaTeX. Paste into Markdown, Overleaf, or Notion. Solve mode also gives the result.
Type a question — "what does this article say?", "summarize the key points" — and the AI answers from the screenshot's text. Multi-turn context, streaming.
8 built-in PII patterns (email, phone, credit card, IPv4/IPv6, MAC, API key, Bearer) plus face detection — all on-device, all free. Pro adds 4-level AI Smart Clean with a line-by-line accept/revert diff preview.
Point SnipMD at a folder and every image dropped in gets OCR'd to Markdown automatically — crash-safe queue, per-file resume, daily summary file, done notification.
Explain in 1 sentence · generate runnable unit test (21+ frameworks: XCTest, JUnit 5, gtest, pytest, Jest, Vitest, xUnit, NUnit, RSpec, Kotest…) · Conventional Commits · add idiomatic doc comments.
One click extracts action items, calendar events, or business-card info. Review sheet lets you edit anything wrong, then push to Apple Reminders, Calendar, or Contacts.
Two App Intents for Shortcuts (capture + OCR, OCR clipboard image) and a headless CLI — snipmd ocr file.png or snipmd capture --mode fullScreen — for scripts and automation.
Arrow, rectangle, text, mosaic, highlight, color picker. Six tools in a clean top bar, plus undo. Copy the annotated PNG or the clean Markdown — your choice.
History syncs across your Apple devices via your private iCloud database — silent push, last-writer-wins, offline queue. Captures on one Mac appear on the other; the iOS companion uploads straight into the same zone.
Share a screenshot library with your team over CloudKit sharing — the system share panel handles invites, every member sees the same captures in their history. Uses the library owner's iCloud storage.
Import a photo and enhance it 4×. On-device Real-ESRGAN (Core ML) never sends the photo anywhere; the cloud variant adds GFPGAN face restoration for best quality.
Drive SnipMD from anything: snipmd:// URL scheme (capture / last / history / settings / ocr-clipboard), headless CLI subcommands (ocr / barcode / links / redact), and Shortcuts actions.
How it works
The shutter is silent on region/window captures. Markdown is ready before your finger leaves the shortcut.
⌥⇧3 ⌥⇧4 ⌥⇧5
Full screen / region / window. Or ⌥⇧I to import an audio or video file for transcription.
Frozen-screen overlay appears. Drag to select. Esc to cancel. Media imports skip this step entirely — drop the file and transcription starts.
Vision OCR runs in < 1s; Whisper transcribes 30s of audio in a few seconds. Annotate, translate, redact, export subtitles, or ask AI — then close.
Privacy & Trust
Changelog
Recent versions, newest first.
snipmd:// URL scheme, CLI barcode/links/redact subcommands, more Shortcuts intents[mm:ss] links seek & play; active row highlights in syncsnipmd ocr / snipmd capture --mode fullScreenRoadmap
Plausible targets, not promises. Dates shift based on what we learn from real users.
Q4 2026 · sync + companions shipped
snipmd:// + CLI)Driven by real user feedback — small, fast.
2027 H1 · structured data
All driven by 1.x Pro feedback. No Pro gating changes.
2027+ · bigger bets
Team workspace and the iOS/Web companions already shipped — see "Now". Heavy bets won't ship without a sustainable Pro base.
Pricing
7-day free trial. Cancel anytime. Pro is purely additive — every Free feature stays free forever.
Everything you need for screenshots — plus a taste of transcription.
Or $59.99/year (save 37%). 7-day free trial. Cloud AI included — no API keys.
FAQ