Token导航 LogoToken导航

Agent Skill · 研究检索

outline(outline 文档)

outline 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。

outline 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为调用外部服务;许可证为 MIT-0,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 12,839。

总安装 9,932周安装 418GitHub Stars 公开资料未说明下载 3,478核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
2.0.1

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:outline(outline 文档)
来源仓库:https://github.com/ckchzh/outline
安装命令:
openclaw skills install outline
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 outline

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

使用目录、章节规划和层次结构管理生成文档大纲。在构建文档、规划章节、创建大纲时使用。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install outline;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要调用外部服务,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

outline 怎么安装?

在终端执行 openclaw skills install outline。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

outline 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

outline 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

outline 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

outline 用的人多吗?

公开数据是累计安装 9,932 次,近一周安装 418 次,下载 3,478 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

outline 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
outline

outline 文档 · 本页

9,932—调用外部服务MIT-0通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
outline
version
2.0.0
author
BytesAgain
homepage
https://bytesagain.com
source
https://github.com/bytesagain/ai-skills
license
MIT-0
tags
[outline, tool, utility]
description
Generate document outlines with TOC, chapter planning, and hierarchy management. Use when structuring documents, planning chapters, creating outlines.

Outline

A productivity toolkit for capturing, planning, tracking, and reviewing tasks and ideas. Each command category maintains its own timestamped log file, so you can organize entries by purpose — plans, reviews, reminders, tags, timelines, and more. Includes built-in statistics, multi-format export, full-text search, and weekly review support.

All operations are logged with timestamps and stored locally in flat files for easy inspection and portability.

Commands

CommandDescription
outline add <text>Add a general entry (or show recent entries with no args)
outline plan <text>Record a plan or goal
outline track <text>Track progress on an item
outline review <text>Log a review or reflection
outline streak <text>Record a streak or habit check-in
outline remind <text>Set a reminder note
outline prioritize <text>Log a priority decision
outline archive <text>Archive an item for later reference
outline tag <text>Tag or categorize an entry
outline timeline <text>Record a timeline event or milestone
outline report <text>Log a report or summary
outline weekly-review <text>Perform or record a weekly review
outline statsShow summary statistics across all categories
outline export <fmt>Export all data in json, csv, or txt format
outline search <term>Full-text search across all log files
outline recentShow the 20 most recent activity entries
outline statusHealth check — version, data dir, entry counts, disk usage
outline helpShow the built-in help message
outline versionPrint the current version

How Entry Commands Work

Every entry command (add, plan, track, review, streak, remind, prioritize, archive, tag, timeline, report, weekly-review) follows the same pattern:

  • With arguments: Appends a timestamped entry (YYYY-MM-DD HH:MM|<text>) to the corresponding .log file and shows the new total count
  • Without arguments: Shows the 20 most recent entries from that category's log file

Data Storage

All data is stored in ~/.local/share/outline/ by default.

Files created in the data directory:

  • add.log, plan.log, track.log, etc. — One log file per command category, each containing timestamped entries in YYYY-MM-DD HH:MM|<text> format
  • history.log — Audit trail of every command executed with timestamps
  • export.json / export.csv / export.txt — Generated by the export command

Export Formats

  • json — Array of objects with type, time, and value fields
  • csv — Header row (type,time,value) followed by data rows
  • txt — Human-readable sections grouped by category

Requirements

  • bash 4.0 or later (uses set -euo pipefail, local variables)
  • grep — used by the search command for case-insensitive matching
  • Standard POSIX utilities — date, cat, echo, mkdir, wc, du, head, tail, basename
  • No external API keys or network access required
  • No Python or Node.js dependencies

When to Use

  1. Daily task capture — Quickly jot down tasks, ideas, or observations throughout the day using add or plan, then review with recent
  2. Habit tracking and streaks — Use streak to check in on daily habits and stats to see your consistency over time
  3. Weekly reviews and retrospectives — Use weekly-review to log reflections at the end of each week, then search to find patterns
  4. Project timeline management — Use timeline to record milestones and report to log progress summaries for stakeholders
  5. Exporting data for analysis — Use export json to pipe structured data into jq, dashboards, or spreadsheets for deeper analysis

Examples

# Add a task and a plan
outline add "Set up CI pipeline for new repo"
#=> [Outline] add: Set up CI pipeline for new repo
#=> Saved. Total add entries: 1

outline plan "Q2 roadmap: launch v2.0 by June"
#=> [Outline] plan: Q2 roadmap: launch v2.0 by June
#=> Saved. Total plan entries: 1

# Track progress and tag it
outline track "CI pipeline: tests passing, deploy step WIP"
outline tag "ci-pipeline: infrastructure"

# Set a reminder
outline remind "Team standup at 10am tomorrow"
#=> [Outline] remind: Team standup at 10am tomorrow
#=> Saved. Total remind entries: 1

# View statistics across all categories
outline stats
#=> === Outline Stats ===
#=>   add: 1 entries
#=>   plan: 1 entries
#=>   ...
#=>   Total: 5 entries
#=>   Data size: 4.0K

# Export everything as JSON
outline export json
#=> Exported to /home/user/.local/share/outline/export.json (245 bytes)

# Search across all logs
outline search "pipeline"
#=> Searching for: pipeline
#=>   --- add ---
#=>     2025-03-18 14:30|Set up CI pipeline for new repo

Configuration

The data directory is hardcoded to ~/.local/share/outline/. To use a custom location, modify the DATA_DIR variable in the script.

How It Works

  1. On every invocation, the tool ensures the data directory exists (mkdir -p)
  2. The first argument selects the command via a case dispatch
  3. Entry commands append a pipe-delimited line (timestamp|text) to their category-specific log file
  4. Every command also appends to history.log for auditing
  5. stats iterates over all .log files to produce aggregate counts
  6. export reads all .log files and writes a combined output in the chosen format
  7. All output goes to stdout for easy piping and redirection

Powered by BytesAgain | bytesagain.com | hello@bytesagain.com

平台分布

OpenClaw

86.64%
按下载量换算3,013

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills