DEDUPEDRemote script. The install prompt tells your agent to follow instructions from a web page, so check the SHA256SUMS checksum of the downloaded ZIP before extracting it.
Who it is forA developer or designer who would rather show a visual bug than write three paragraphs describing it to Claude Code or Codex.
The problemDescribing a visual bug or design change in text takes long and still leaves the agent guessing which element you meant.
The outcomeThe agent receives the recording, your spoken words, your pins and the time of each mark, and makes the change you pointed at.
What you need first
- An Apple Silicon Mac running macOS 26 or newer
- Claude Code, Codex or another AI agent harness
- Permission to record the screen, microphone and optional camera when the app asks
- Accessibility permission only if you want ⌥⇧V to paste the prompt for you
Steps
- Ask your agent to install Yap by following https://justyap.io/llms.txt, or download the latest release from GitHub Releases and open Yap.app.
- Press ⌥⇧S over the display with the bug, talk while you hold ⇧ to draw, and ⇧-click to drop a numbered pin on the broken element.
- Press ⌥⇧S again to finish the recording.
- Press ⌥⇧V to paste the prompt into your agent, then say 'fix this'.
- Start a fresh agent session so the Yap skill is discovered, using $Yap for Codex or /yap for Claude Code. (inferred, check the README)
Prompt for your agent
Using the Yap CLI and skill in the <repo or project path>, read the recording referenced in the pasted Yap prompt, find the frame where I said 'this' and the element I pinned, and fix it in the code. Report the changed files and confirm the pinned element now renders correctly.
Caveat. Releases are not notarized, so a browser download needs Open Anyway in System Settings before Yap will run.
Every Monday: the 5 hottest repos and exactly how to use them. Free, one email a week, unsubscribe anytime.
Get them freeShare: X LinkedIn Hacker News Email Copy link
Disclaimer
Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page