ChosenXu

Raindrop Bookmark Organizer — Git skill for Claude Code

Git community

Agent Skills-compatible skill: batch-organize a Raindrop.io library — structured notes, three-axis controlled tags, undo snapshots & verified writes.

How to install Raindrop Bookmark Organizer

This entry records only its repository, not the path inside it, so there is no exact command to give. Open ChosenXu/raindrop-bookmark-organizer and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Raindrop Bookmark Organizer does

Agent Skills-compatible skill: batch-organize a Raindrop.io library — structured notes, three-axis controlled tags, undo snapshots & verified writes. Works with Claude Code, Codex CLI, Gemini CLI, GitHub Copilot, Cursor & WorkBuddy.

Alternatives in Git

Epic Sync — Sync epic issue bodies, labels, and local coordination snapshots from GitHub 243.5k ★
Vibe Log — by Vibe-Log - Analyzes your Claude Code prompts locally (using CC), provides intelligent session analysis and 306 ★
Eval Merge — /eval-merge — Three-Way Version Merge 216 ★

README

Raindrop Bookmark Organizer

English | [简体中文](README.zh-CN.md)

An [Agent Skills](https://agentskills.io)-compatible skill that batch-organizes a [Raindrop.io](https://raindrop.io/) library with a proven three-step workflow — **name (optional) → annotate → tag**. Works in any agent that reads the Agent Skills standard (Claude Code, Codex CLI, Gemini CLI, GitHub Copilot, Cursor, WorkBuddy…). Battle-tested end-to-end on a real 1,101-bookmark library: 37 batches, 100% tagged + annotated, zero structural violations.

What it does

For every untagged bookmark it produces:

Note — 1–2 sentences: what it is + when you'll reach for it (40–80 characters in Chinese, 30–60 words in English)
Tags — up to 6, selected verbatim from a controlled three-axis vocabulary:
- Domain (1–2): 10 categories ≈ 37 leaves — Dev / Design / AI / Productivity / Hardware / Media & Culture / Learning / Science / News / Utilities
- Type (exactly 1): 16 types in 5 semantic groups — Assets / Interactive / Content / Discovery / Identity
- Status (0–2, off by default): to-read · starred · free · oss · paid · zh · en
Title rewrite — conservative, opt-in, default OFF; garbage titles only

Design principle: **metadata first**. Raindrop's list API returns title/link/domain/excerpt in bulk — no webpage fetching needed for ~90% of bookmarks. Deep-fetch (`fetch_bookmark_content`) is reserved for the low-confidence subset.

Tag output language is configurable: the skeleton ships with Simplified Chinese and English canonical columns, and the custom vocabulary file can rename any tag or switch the tag language per library.

Highlights

Metadata first, fetch only when unsure — bulk list API returns title/link/domain/excerpt/type per bookmark; webpage fetching is a tier-2 fallback for the low-confidence subset, not the default path.
Structured, reusable, and on-track — tags are selected verbatim from a controlled two-layer vocabulary (universal skeleto