← all repositories
StashDock/StashDock

A local library for the stuff you keep feeding to AI

StashDock gathers links, files and prompts into one local, visual library you can drag into any AI conversation.

StashDock
Too new to rank. Velocity needs a few days of history.
not enough history yet
star history

What it does StashDock is a personal context library that lives on your machine. You collect webpages, local files and prompt/rule notes as visual cards, organize them with tags and favorites, and then pull them into AI chats — via a ChatGPT sidebar plugin, a DeepSeek Harness plugin, or a standalone desktop app with drag-and-drop. A Chrome extension captures pages as you browse, including readable content and a cover image, and you review the capture before it’s saved.

The interesting bit The library is exposed through MCP, so AI agents can do more than read: they can recommend saved material for the current task, capture reusable notes from the conversation back into the library, and propose tag changes — which you review before they’re applied. It’s a two-way context loop, not just a clipboard.

Key highlights

  • Local-first: references stored on-device in SQLite with FTS5 full-text search; no cloud sync between machines.
  • Imports bookmarks from Chrome, Edge, Firefox and Safari without an extension.
  • Desktop app (Tauri 2/Rust, Windows and macOS) with a floating folder for quick search and drag-into-chat.
  • Shared React/TypeScript interface reused across desktop and AI plugins, so both read the same library.
  • AGPL v3 licensed.

Caveats

  • No automatic sync between libraries on different computers — by design, but worth knowing.
  • The ChatGPT plugin and DeepSeek Harness plugin are the named AI integrations; other AI apps work only via generic copy/drag from Desktop.
  • First launch triggers unsigned-app verification prompts on both Windows and macOS.

Verdict Worth a look if you constantly re-paste the same links, files and prompts into AI chats and want that tamed locally. If your AI tooling isn’t ChatGPT or DeepSeek Harness, you’re mostly getting a nice drag-and-drop reference manager.

Frequently asked

What is StashDock/StashDock?
StashDock gathers links, files and prompts into one local, visual library you can drag into any AI conversation.
Is StashDock open source?
Yes — StashDock/StashDock is open source, released under the AGPL-3.0 license.
What language is StashDock written in?
StashDock/StashDock is primarily written in HTML.
How popular is StashDock?
StashDock/StashDock has 576 stars on GitHub.
Where can I find StashDock?
StashDock/StashDock is on GitHub at https://github.com/StashDock/StashDock.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.