Token导航 LogoToken导航TokenDH.com
研究检索敏感数据clawhub未标认证来源可访问clear审计通过

karakeep-app卡拉 OK 应用程序

Agent Skill

karakeep-app 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

4,469

周安装

190

GitHub Stars

1

下载量

1,566
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:karakeep-app(卡拉 OK 应用程序)
来源仓库:https://github.com/karakeep/karakeep-app
安装命令:
openclaw skills install karakeep-app
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 karakeep-app

简介

karakeep-app 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。

  • 适用于书签管理和信息检索等研究类任务。
  • 通过 openclaw skills install karakeep-app 安装使用。
  • 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。
  • 可结合来源仓库和原始 README 继续核验具体用法。

SKILL.md

name
karakeep
description
Official skill for how to use karakeep (the bookmark manager) and interact with it programmatically.
metadata
tags
bookmarks, bookmark manager, 2nd brain, productivity
openclaw
envVars
required
true
description
The API key for your Karakeep instance
required
false
description
The server address for your Karakeep instance.
requires
env
bins
homepage
https://karakeep.app
links
repository
https://github.com/karakeep-app/karakeep
documentation
https://docs.karakeep.app
emoji
📦
cliHelp
karakeep --help
install
package
@karakeep/cli
bins
[karakeep]

Karakeep

Karakeep is an open source self-hosted bookmark manager for collecting, organizing, and searching content. This skill covers the core concepts and how to interact with Karakeep via the CLI.

When to use

Use this skill when the user wants to interact with their Karakeep instance (adding bookmarks, managing lists/tags, searching, etc.).

Core Concepts

Bookmarks

  • Bookmarks: Core entity in Karakeep. Can be one of links, text or media.

- Links: Save URLs — Karakeep auto-fetches title, description, image, screenshot, and full-page archive. - Text: Quick notes or text snippets stored as bookmarks. - Media: Images and PDFs uploaded directly.

  • Favorites: Star bookmarks for quick access.
  • Archiving: Hide bookmarks from the homepage while keeping them searchable.
  • Notes: Attach personal context notes to any bookmark.
  • Highlights: Save quotes, summaries, or TODOs while reading — searchable across all bookmarks.

Lists

  • Manual lists: Curated collections organized by project or topic. Can be private or public.
  • Smart lists: Auto-updating lists powered by search queries (e.g., #ai -archived).
  • Collaboration: Invite editors (can add bookmarks) or viewers (read-only) to a list.

Tags

Lightweight labels for any bookmark (topics, sources, workflow states). Multiple tags per bookmark, tags travel with bookmarks across lists. AI can auto-generate tags when configured.

Search Query Language

Karakeep has a powerful search query language for finding the right bookmarks. It supports full-text search, boolean logic, qualifiers, and more.

Basic Syntax

  • Spaces between conditions act as implicit AND.
  • Use and / or keywords for explicit boolean logic.
  • Prefix any qualifier with - or ! to negate it (e.g., -is:archived, !is:fav).
  • Use parentheses () for grouping (note: groups themselves can't be negated).
  • Any text not part of a qualifier is treated as full-text search (e.g., machine learning is:fav).

Qualifiers

QualifierDescriptionExample
is:favFavorited bookmarksis:fav
is:archivedArchived bookmarks-is:archived
is:taggedBookmarks with one or more tagsis:tagged
is:inlistBookmarks in one or more listsis:inlist
is:linkLink bookmarksis:link
is:textText/note bookmarksis:text
is:mediaMedia bookmarks (images/PDFs)is:media
is:brokenBookmarks with failed crawls or non-2xx status codesis:broken
url:<value>Match URL substringurl:github.com
title:<value>Match title substring (supports quoted strings)title:rust, title:"my title"
#<tag> or tag:<tag>Match bookmarks with specific tag (supports quoted strings)#important, tag:"work in progress"
list:<name>Match bookmarks in a specific list (supports quoted strings)list:reading, list:"to review"
after:<date>Created on or after date (YYYY-MM-DD)after:2024-01-01
before:<date>Created on or before date (YYYY-MM-DD)before:2024-12-31
age:<time-range>Filter by creation age. Use < / > for max/min age. Units: d (days), w (weeks), m (months), y (years)age:<1d, age:>2w, age:<6m
feed:<name>Bookmarks imported from a specific RSS feedfeed:Hackernews
source:<value>Match by capture source. Values: api, web, cli, mobile, extension, singlefile, rss, importsource:rss, -source:web

Examples

# Favorited bookmarks from 2024 tagged "important"
is:fav after:2024-01-01 before:2024-12-31 #important

# Archived bookmarks in "reading" list or tagged "work"
is:archived and (list:reading or #work)

# Untagged or unorganized bookmarks
-is:tagged or -is:inlist

# Recent bookmarks from the last week
age:<1w

# Full-text search combined with qualifiers
machine learning is:fav -is:archived

RSS Feeds

Karakeep can also be used to consume RSS feeds, but also can itself act as an RSS feed publisher.

  • Publishing: Export any list as an RSS feed with a unique token.
  • Consuming: Auto-monitor external RSS feeds and create bookmarks from new items (hourly, with duplicate detection).

Automation

  • Rule Engine: If-this-then-that rules to auto-tag, favorite, or route bookmarks to lists.
  • Webhooks: Subscribe to bookmark events (add/update/archive).

Interacting with Karakeep via the CLI

Installation

npm install -g @karakeep/cli

Or via Docker:

docker run --rm ghcr.io/karakeep-app/karakeep-cli:release --help

Authentication

The CLI requires an API key and server address. Get the API key from your Karakeep instance's settings page.

Option 1 — Environment variables (recommended):

export KARAKEEP_API_KEY="your-api-key"

# If self-hosted, pass the server address as well. It defaults to the cloud instance if not set:
export KARAKEEP_SERVER_ADDR="https://cloud.karakeep.com"

Option 2 — CLI flags:

karakeep --api-key <key> --server-addr <addr> <command>

Verify authentication:

karakeep whoami

Bookmark Commands

Run karakeep --help to see all available commands, but the most important ones are:

# Add a link bookmark
karakeep bookmarks add --link "https://example.com"

# Add a link with tags and to a specific list
karakeep bookmarks add --link "https://example.com" --tag-name "reading" --list-id <list-id>

# Add a text bookmark
karakeep bookmarks add --note "Remember to review the PR"

# Get bookmark details
karakeep bookmarks get <bookmark-id>
karakeep bookmarks get <bookmark-id> --include-content

# Update a bookmark
karakeep bookmarks update <bookmark-id> --title "New Title"
karakeep bookmarks update <bookmark-id> --archive
karakeep bookmarks update <bookmark-id> --favourite
karakeep bookmarks update-tags <bookmark-id> --add-tag "important"
karakeep bookmarks update-tags <bookmark-id> --remove-tag "old-tag"

# List management
karakeep lists list
karakeep lists get --list <list-id>
karakeep lists add-bookmark --list <list-id> --bookmark <bookmark-id>
karakeep lists remove-bookmark --list <list-id> --bookmark <bookmark-id>
karakeep lists delete <list-id>


# List all bookmarks
karakeep bookmarks list

# Search bookmarks
karakeep bookmarks search "is:fav #work"
karakeep bookmarks search "rust" --limit 10 --sort-order relevance
karakeep bookmarks search "is:tagged" --all   # paginate through all results

# Delete a bookmark
karakeep bookmarks delete <bookmark-id>

You can always pass --json to get raw JSON output instead of pretty-printed output.

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

91.02%
按下载量换算1,425

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。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