Raindrop Sync

Sync and process bookmarks from Raindrop.io. Use when fetching new bookmarks, analyzing saved content, or syncing bookmarks to knowledge base. Triggers on "r...

Security Scan

The skill's code and instructions match the stated Raindrop syncing purpose, but the package metadata fails to declare the required RAINDROP_TOKEN credential and the runtime instructions ask the agent to read/write local secret and memory files — an inconsistency that could hide credential requirements and broaden data access.

Purpose & Capability

The script fetches bookmarks from api.raindrop.io and the SKILL.md describes fetching, filtering, and storing bookmarks — this aligns with the skill name and description. However, the registry metadata lists no required environment variables or primary credential while the SKILL.md and script both expect RAINDROP_TOKEN; that mismatch is unexpected.

Instruction Scope

SKILL.md instructs sourcing a local secrets file (.secrets/raindrop.env) and mentions reading/writing memory/kb-index.json and memory/knowledge-base.md and calling an external 'web_fetch' step. The included script only performs API fetch and outputs JSON (stdout or file). The extra workflow steps are outside the script but instruct the agent to access local files; this expands the runtime scope beyond a pure fetcher but is coherent with syncing/processing bookmarks.

Install Mechanism

No install spec (instruction-only plus a small included script). Nothing is downloaded or written to system install locations by the skill itself — this is the lowest-risk install pattern.

Credentials

The script and SKILL.md require a Raindrop API token (RAINDROP_TOKEN) but the skill metadata declares no required env vars or primary credential. Omitting a required secret from the manifest can hide the fact that a credential is needed and bypass platform permission prompts. SKILL.md also recommends storing the token in a plaintext file (.secrets/raindrop.env), which has local security implications.

Persistence & Privilege

always is false, the skill does not request persistent/autonomous elevation beyond the platform default, and it does not modify other skills or system-wide configs in its files.

What to consider before installing

This skill appears to do what it claims (fetch Raindrop bookmarks) and the included script is small and readable, but take these precautions before installing: 1) The registry metadata does NOT declare RAINDROP_TOKEN even though the script and SKILL.md require it — ensure you understand where/when you'll supply the token and prefer a platform-managed secret rather than a plaintext file in .secrets/. 2) If you must store the token on disk, keep it out of source control and restrict file permissions. 3) Confirm the only network calls are to api.raindrop.io (the script uses that URL). 4) Be aware the workflow instructs reading/writing memory/kb-index.json and appending to knowledge-base files — verify those paths do not expose sensitive data you don't want the skill to access. 5) If you want the platform to surface a permission prompt, ask the skill author to add RAINDROP_TOKEN to requires.env/primaryEnv in the metadata. If anything about where the token is stored or how the agent will be allowed to run the workflow is unclear, treat this as a potential privacy risk and inspect or run the script locally first.

Raindrop Bookmark Sync

Fetch bookmarks from Raindrop.io API and process them for knowledge base integration.

Setup

1. Get API token from https://app.raindrop.io/settings/integrations
2. Create test token (read-only is fine)
3. Save to .secrets/raindrop.env:

   RAINDROP_TOKEN=your_token_here
   

Usage

Fetch new bookmarks

source .secrets/raindrop.env
python3 skills/raindrop/scripts/fetch.py --since 24h

Fetch from specific collection

python3 skills/raindrop/scripts/fetch.py --collection 12345678

Process and add to knowledge base

python3 skills/raindrop/scripts/fetch.py --since 24h --output /tmp/raindrop-new.json
# Then process each item with web_fetch and add to memory/knowledge-base.md

API Reference

• Base URL: https://api.raindrop.io/rest/v1
• Auth: Bearer token in header
• Rate limit: 120 req/min

Key Endpoints

• GET /raindrops/{collectionId} — List bookmarks (use 0 for all)
• GET /collections — List collections
• GET /raindrop/{id} — Single bookmark details

Bookmark Object

{
  "_id": 123456,
  "title": "Article Title",
  "link": "https://example.com/article",
  "excerpt": "Short description...",
  "tags": ["tag1", "tag2"],
  "created": "2026-02-15T10:00:00Z",
  "collection": {"$id": 12345678}
}

Workflow

1. Fetch — Get new bookmarks since last sync
2. Filter — Skip already-processed URLs (check memory/kb-index.json)
3. Extract — Use web_fetch to get content
4. Analyze — Summarize and tag
5. Store — Append to memory/knowledge-base.md
6. Update index — Add URL to memory/kb-index.json

Cron Integration

Add to heartbeat or cron for automatic sync:

每天检查一次 Raindrop 新书签，处理后存入知识库