Scrawl records your screen, voice, clicks and drawings, then packs them into one compact capture you paste into your AI coding agent. Press ⌥⇧S, talk, click and draw. Frames are taken only when something happens, speech is transcribed on your Mac, and each frame is cropped to where you worked. Works with iPhone mirroring and ScrawlKit for iOS apps. Open source (MIT). macOS 15+, Apple silicon.
Ever typed out "the button is slightly off, no, the other button, near the top-left, kind of under the header…" to an AI coding agent? I have, way too often. Words are a bad way to describe pixels. That's why I hunted Scrawl.
Point, talk, draw, paste. Press ⌥⇧S, explain the bug out loud while you click and sketch on your screen, then paste the whole thing into your agent. Scrawl only grabs a frame when something actually happens, and each frame keeps the sentence you said at that moment. Your agent sees what you saw and hears what you meant.
What won me over:
🔒 Transcription runs on your device, so nothing gets sent to a server
✂️ Auto-crops to the window you were working in
📱 Works with iPhone too (mirroring, plus ScrawlKit for iOS apps)
🆓 Open source (MIT), and one line installs it:
brew install --cask kitlangton/tap/scrawl
Full disclosure: I'm just the hunter. @kitlangton built it and can answer any question about how it works.
Give it a spin and tell me which agent you paired it with. I'm curious who's drawing for Claude, Cursor or something else entirely 👇
“The other button, near the top-left, kind of under the header” is painfully familiar — I've wasted so many prompts trying to describe a visual bug in words. Recording voice, clicks and drawings into one capture makes a lot more sense, and I like that frames are only taken when something actually happens, transcription stays on-device, and it auto-crops to the window you were working in instead of your whole screen.
Describing a UI bug in words is the worst part of building with Claude. Our app is on iOS, so the iPhone mirroring support is what sells it for us.
About Scrawl on Product Hunt
“Show your AI agent the problem with a visual frame”
Scrawl launched on Product Hunt on November 3rd, 2026 and earned 75 upvotes and 3 comments, placing #24 on the daily leaderboard. Scrawl records your screen, voice, clicks and drawings, then packs them into one compact capture you paste into your AI coding agent. Press ⌥⇧S, talk, click and draw. Frames are taken only when something happens, speech is transcribed on your Mac, and each frame is cropped to where you worked. Works with iPhone mirroring and ScrawlKit for iOS apps. Open source (MIT). macOS 15+, Apple silicon.
Scrawl was featured in Mac (103.8k followers), Developer Tools (522.4k followers), Artificial Intelligence (482.8k followers) and GitHub (41.5k followers) on Product Hunt. Together, these topics include over 270.8k products, making this a competitive space to launch in.
Who hunted Scrawl?
Scrawl was hunted by Mayur Sojitra. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
Want to see how Scrawl stacked up against nearby launches in real time? Check out the live launch dashboard for upvote speed charts, proximity comparisons, and more analytics.
Mayur Sojitra
Hey everyone! I'm Mayur 👋
Ever typed out "the button is slightly off, no, the other button, near the top-left, kind of under the header…" to an AI coding agent? I have, way too often. Words are a bad way to describe pixels. That's why I hunted Scrawl.
Point, talk, draw, paste. Press ⌥⇧S, explain the bug out loud while you click and sketch on your screen, then paste the whole thing into your agent. Scrawl only grabs a frame when something actually happens, and each frame keeps the sentence you said at that moment. Your agent sees what you saw and hears what you meant.
What won me over:
🔒 Transcription runs on your device, so nothing gets sent to a server
✂️ Auto-crops to the window you were working in
📱 Works with iPhone too (mirroring, plus ScrawlKit for iOS apps)
🆓 Open source (MIT), and one line installs it:
Full disclosure: I'm just the hunter. @kitlangton built it and can answer any question about how it works.
Give it a spin and tell me which agent you paired it with. I'm curious who's drawing for Claude, Cursor or something else entirely 👇