explaingit

thapasamir/pastry

Analysis updated 2026-05-18

8SwiftAudience · vibe coderSetup · easy

TLDR

A local Mac clipboard manager that lets you search your copy history by meaning instead of exact words.

Mindmap

mindmap
  root((Pastry))
    What it does
      Semantic clipboard search
      Local embedding matching
      Filter by content kind
    Tech stack
      Swift menu bar app
      Python engine
      ChromaDB search index
      SQLite storage
    Use cases
      Recall old clips by meaning
      Store screenshots and text
      Pin reusable snippets
    Audience
      Mac users
      Vibe coders
      Privacy focused users

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Find a clip you copied earlier by describing what it was about instead of remembering exact words.

USE CASE 2

Keep a searchable local history of copied text, links, code, and screenshots without any cloud sync.

USE CASE 3

Pin frequently reused snippets like API keys or addresses so they stay easy to grab again.

What is it built with?

SwiftSwiftUIPythonChromaDBSQLite

How does it compare?

thapasamir/pastry98przem/mistgistya/ox0badf00d
Stars888
LanguageSwiftSwiftSwift
Setup difficultyeasyeasymoderate
Complexity3/54/5
Audiencevibe codergeneraldeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · easy Time to first run · 5min

Download the DMG and drag to Applications, or run install.sh from source, Apple Silicon and macOS 14+ only.

In plain English

Pastry is a clipboard manager for Mac computers that lets you search everything you have ever copied by describing what it means, not just by typing the exact words it contains. Press a keyboard shortcut and type something like "the api key from this morning" or "mike's address," and Pastry finds the right clip even if your search does not share a single word with it. It does this using a small local model that turns text into meaning based comparisons, so it does not need the internet to work. Every time you copy something, whether it is text, a screenshot, or an image, Pastry saves it automatically and remembers which app it came from. It can tell the difference between plain text, links, emails, colors, and code, and shows each type a little differently, for example a hex color gets a small preview swatch. If you copy the same thing twice, it just moves back to the top instead of creating a duplicate. You can pin important clips so they never get deleted, and set rules for how long regular clips and images stick around before being cleaned up automatically. Everything runs entirely on your own computer. There are no accounts, no network calls, and no data sent anywhere. All your clipboard history lives in a folder on your Mac, and deleting that folder wipes everything. Content that password managers mark as sensitive is skipped so it never gets saved in the first place. Under the hood, Pastry pairs a Swift app you see and interact with, with a small Python program running quietly in the background that watches your clipboard and handles the searching. The easiest way to use it is downloading the ready made app from the project's releases page and dragging it into your Applications folder, no coding tools required. Developers who want to build it from source can run a single install script instead. It currently works only on newer Apple Silicon Macs running a recent version of macOS.

Copy-paste prompts

Prompt 1
Walk me through installing Pastry on my Mac using the downloadable app, step by step.
Prompt 2
Explain how Pastry's semantic search works and why it can match a query that shares no words with the clip.
Prompt 3
Show me how to run Pastry from source using the install.sh script on an Intel Mac.
Prompt 4
Help me understand the IPC protocol Pastry's engine uses so I can build a small script that talks to it.

Frequently asked questions

What is pastry?

A local Mac clipboard manager that lets you search your copy history by meaning instead of exact words.

What language is pastry written in?

Mainly Swift. The stack also includes Swift, SwiftUI, Python.

How hard is pastry to set up?

Setup difficulty is rated easy, with roughly 5min to a first successful run.

Who is pastry for?

Mainly vibe coder.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.