clipboard-tool

Project Overview

Clipboard Tool is a multi-user clipboard manager that runs as a tools-center platform tool, with a Node.js backend that has zero third-party dependencies. It offers a universal entry point: paste anything and it is automatically recognized and saved; URLs become link items, other text becomes text items, dropped or Ctrl+V files become file items, and copied images can be stored directly. The UI uses a dark Neumorphism style with a black-gold luxury color scheme, a minimal wall layout on the home page, and it supports multiple users, tagging, pinyin-initial search, smart sorting, and WebDAV backup and multi-device sync. It solves the everyday problems of scattered copied content, hard-to-search history, no cross-device sharing, and easily lost important clips.

Key Features

• Single universal entry point: paste auto-detects the type (URL becomes a link, other content becomes text), drag-and-drop or file picking creates file items, and Ctrl+V pastes images or files; detected copied content automatically opens a large window
• One-click copy and double-click editing: text and links copy their content in one click, images are written straight to the system clipboard, and other files download on click
• Rich-text dual-format copy: content copied from web pages or Word is detected via text/html; the card splits left and right so the left column copies plain text and the right column pastes into Word or Feishu with formatting preserved; Word private properties and document-level settings are preserved at the string level; the edit dialog offers a Clear Format action to convert to plain text
• Smart sorting: starred items float to the top, then by usage count, then by most recent update, so more frequently clicked items rank higher
• Pinyin-initial search: a built-in mapping of 3755 common characters lets you type sf to find the item named 身份
• Tagging: pick existing tags or type new ones, filter the list by tag, and tags wrap automatically so all remain visible when numerous
• Multi-user: a dual-name model with an account name (immutable after creation, serving as the identity key for cross-device migration and sync) and a display name (freely changeable, display only); supports password-free frictionless entry or an optional password lock, with persistent sessions (stay logged in after restart, valid for 30 days)
• WebDAV backup and sync: one-way full backup and two-way merge sync, with a tombstone mechanism to prevent deleted items from reviving, full clears that do not propagate deletions, an automatic sync timer defaulting to 12 hours, optional syncing of file blobs, and a blank address defaulting to the LAN server; snapshots are addressed by account name, so a new device that creates the same account name, configures WebDAV, and syncs pulls back all data
• Rolling archive: the active area caps at 500 items per user, with overflow moved to the archive losslessly; an Include Archive option searches history, the archive participates in full WebDAV backups, and cards can be restored or deleted
• Local backup: export everything (including the archive) to JSON in one click and import with merge (keeping the newer of duplicate ids), independent of WebDAV
• Other useful features: saving links automatically strips UTM and other tracking parameters, tags can be renamed or deleted in settings, JSON text items offer one-click formatting preview, and the top bar toggles between color and grayscale palettes
• Security measures: UUID whitelisting against path traversal, atomic writes, login rate limiting, an upload blacklist (rejecting executables and scripts), and forced attachment downloads

Target Audience

• Users who want to share clipboard content across multiple devices
• Users who often copy rich text from web pages or Word and want formatting preserved
• Users who want to organize and long-term keep scattered copied text, links, images, and files
• Users who value data ownership and want self-hosted WebDAV backup
• Developers who prefer self-hosted, single-file, zero-dependency tools

Quick Start

1. Make sure Node.js 22.7 or later is installed.
2. Run node server.mjs [port] in the project root to start the service (default port 8130); the data directory defaults to ./.data and can be overridden with the CAP_STORAGE_DIR environment variable.
3. Open the corresponding address in your browser: pasted content is auto-detected and saved, copied images or files also open a large window automatically, and JSON text items can be previewed with one-click formatting.
4. For the local server edition, unzip and double-click delivery/server/start-server.cmd (requires Node.js 22.7+, zero dependencies and no npm install); for the platform edition, place the whole clipboard directory into the tools-center tools mount directory and let the platform rescan.
5. For multi-device sync, open Data Management then WebDAV, enter the same address and use the same account name, and click One-Click Sync on each device; to let both devices pull from each other, sync first on the device holding the images or files to push those blobs to the cloud.
6. On startup the first console line prints a version fingerprint such as clipboard v0.7.6 plus a commit id, so you can confirm which version is running.

Core Advantages

• Zero third-party dependencies: the backend uses only Node.js built-in http and fetch, making deployment simple
• Universal entry point: paste-to-save with automatic detection of text, links, images, and files, requiring almost no manual classification
• Rich-text fidelity: content copied from web pages and Word keeps its formatting, and plain text is one click away
• Data ownership: self-hosted backup and sync over standard WebDAV, with snapshots addressed by account name for zero-config device migration
• Single-file frontend: native HTML and JS with no framework, giving a polished UI that is easy to customize

System Requirements

• OS: macOS, Linux, Windows, and any platform that can run Node.js
• Architecture: cross-platform (depends on the Node.js runtime)
• Additional dependencies: Node.js 22.7 or later; the server has zero third-party dependencies and needs no npm install; multi-device sync requires your own WebDAV service

Notes

• The /tmp paths in the docs are Unix-style; on Windows use absolute paths such as C:/Temp/clip-test instead.
• When running smoke or WebDAV tests, use a separate data directory and never run them against real data.
• The exe desktop edition (clipboard-exe directory) is a C# WPF rewrite that is currently planned or in development and may require the .NET 9 Desktop Runtime.
• The Sync File Blobs option in WebDAV sync is enabled by default: it only decides whether local blobs are uploaded, while pulling missing blobs from the cloud always runs, so a new device can download files even if the option is off.