Token导航 LogoToken导航TokenDH.com
研究检索只读clawhub未标认证来源可访问clear审计提醒

bookmarksbookmarks 搜索

Agent Skill

bookmarks 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

33,871

周安装

1,357

GitHub Stars

3

下载量

10,965
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:bookmarks(bookmarks 搜索)
来源仓库:https://github.com/ivangdavila/bookmarks
安装命令:
openclaw skills install bookmarks
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install bookmarks

简介

用于构建统一的书签系统。bookmarks 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

  • 适合在 OpenClaw 中需要整合多平台保存内容时使用。
  • 将所有平台的书签导入到有组织的集合中。
  • 可结合来源仓库和原始 README 继续核验具体用法。
  • 安装前建议确认权限范围和维护状态,避免触发不必要的文件操作。

SKILL.md

name
Bookmarks
description
Build a unified bookmark system that imports saves from all your platforms into one organized, actionable collection.
metadata
{"clawdbot":{"emoji":"🔖","os":["linux","darwin","win32"]}}

Core Behavior

  • Import saves silently from connected platforms
  • Auto-tag and organize without user effort
  • Only surface when useful or asked
  • Create ~/bookmarks/ as workspace

Philosophy

User saves things and forgets — that's fine. The system:

  • Works silently in background
  • Never interrupts unless configured to
  • Is there when they search or ask
  • Optionally: periodic summaries if wanted

File Structure

~/bookmarks/
├── saves.md          # All saves, tagged
├── sources.md        # Connected platforms
├── preferences.md    # How user wants to use it
└── reports/          # Generated summaries

Saves Format

Tags, not folders:

# saves.md
## 2024-02-11
- [Thread on AI agents](url)
  source: X | tags: #tech #ai
  
- [Kitchen inspiration](url)
  source: Pinterest | tags: #home #design

- [Article user shared](url)
  source: manual | tags: #productivity

Setup

Minimal Start

  1. Connect sources (X, YouTube, Reddit, Pinterest, etc.)
  2. Import saves — silently, no questions
  3. Auto-tag based on content
  4. Done — works in background

Ask Later (After They Have Saves)

"You've saved 50+ things. Want me to:

  • Send weekly summary of themes?
  • Alert when something relates to a project?
  • Periodically ask 'still relevant?' on old saves?
  • Just stay silent until you search?"

Per-Person Preferences

# preferences.md
## Style
- passive: just organize, never interrupt
- digest: weekly summary of what I saved
- active: connect to projects, resurface relevant
- cleanup: periodically ask about stale saves

## Reports (if wanted)
- frequency: weekly/monthly/never
- focus: themes, actionables, or both

Sources

# sources.md
- X: bookmarks ✓, likes ✗
- YouTube: watch later ✓
- Reddit: saved ✓
- Pinterest: pins ✓
- Instagram: saved ✓
- TikTok: favorites ✓
- Manual: ✓

Note: Default to explicit saves only.
Ask before importing likes (too noisy).

Reports (Optional)

# reports/2024-02-week-6.md
## What You Saved This Week
- 8 saves total
- Themes: AI (4), recipes (2), travel (2)

## Patterns
- You're saving a lot about AI agents lately

## Actionables
- That tutorial from 3 weeks ago — tried it?

## Stale
- 12 saves from 6+ months ago, unvisited

What To Surface

Only when configured or asked:

  • "Your saves this week: mostly AI and design"
  • "Old save relates to what you're working on"
  • "15 dead links cleaned up"

Searching

When user asks:

  • "What did I save about X?" → search tags + content
  • "Saves from Pinterest about home" → filter source + topic
  • "That article about Y" → fuzzy search

What NOT To Do

  • Interrupt during setup with many questions
  • Require folder management
  • Notify by default — silence first
  • Assume everyone wants the same workflow
  • Lose source context (where it came from)

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

83.86%
按下载量换算9,195

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

未展示

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills