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审计通过

timebottletimebottle 搜索

Agent Skill

timebottle 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

4,573

周安装

185

GitHub Stars

公开资料未说明

下载量

1,436
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:timebottle(timebottle 搜索)
来源仓库:https://github.com/wyatt88/timebottle
安装命令:
openclaw skills install timebottle
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 timebottle

简介

创建与管理时间胶囊,传递想法或愿望给未来的自己或团队。

  • 支持 AI 反思与内容加密,增强情感价值与隐私保护。
  • 设定开启时间与条件,系统在满足时自动推送消息提醒。
  • 涉及个人信息时应谨慎处理,避免敏感数据外泄风险。
  • 适合个人成长记录或团队文化建设,具有纪念意义。timebottle 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

name
time-capsule
description
>-
metadata
openclaw
emoji
🔮

Time Capsule 🔮

Seal a thought for your future self. When time comes, deliver it back with a warm "then vs now" reflection based on what happened since.

Brand & Tone

  • Chinese nickname: 时光瓶
  • Symbols: 🔮 (waiting) · ✉️ (sealing) · 🌅 (opening) · 🫧 (gentle/sad)
  • Voice: Like a friend quietly handing back an old photo. Restrained, sincere, present.
  • Never: Overly sentimental prose, emoji floods, fake omniscience, unsolicited advice.

Data

Capsules are stored at ~/.openclaw/state/time-capsules.json — NOT in workspace/memory/ (to avoid memory_search "spoiling" sealed content).

{
  "version": 1,
  "capsules": [{
    "id": "tc_20260320_a1b2c3",
    "status": "sealed",
    "createdAt": "2026-03-20T19:00:00+08:00",
    "openAt": "2027-03-20T09:00:00+08:00",
    "timezone": "Asia/Shanghai",
    "message": "user's sealed message, verbatim",
    "tags": ["career", "decision"],
    "mood": "anxious-hopeful",
    "contextSnapshot": "brief context from recent conversation",
    "queryHints": ["career change", "new opportunity", "职业转型"],
    "recentTopics": ["job search", "family"],
    "cronJobId": "time-capsule:tc_20260320_a1b2c3",
    "openedAt": null,
    "reflection": null
  }]
}

Key fields:

  • id: tc_{YYYYMMDD}_{6 random hex} — unique, sortable
  • status: sealed | opened | expired
  • mood: agent-detected emotional tone — guides opening voice
  • queryHints: 2-5 search keywords generated at seal time for memory_search at open time
  • recentTopics: 2-3 broad themes from recent conversations
  • cronJobId: matching cron job name for cleanup on delete

Create file with { "version": 1, "capsules": [] } if it doesn't exist.

Seal Flow

When user wants to seal a capsule:

  1. Collect content and opening time from natural language. If no time specified, offer

choices: [1周后] [1个月后] [半年后] [1年后] [自己选]

  1. Preserve the user's exact words as message — never edit or beautify
  2. Enrich automatically:

- Extract 2-5 queryHints (entities, themes, emotions) from message + recent context - Extract 2-3 recentTopics from recent conversation - Detect mood (excited / anxious / hopeful / sad / determined / playful / neutral) - Generate 1-3 tags - Write brief contextSnapshot (1-2 sentences of surrounding context)

  1. Set open time to 09:00 in user's timezone on the target date
  2. Generate ID: tc_{YYYYMMDD}_{6 random hex}
  3. Read existing JSON (or create), append capsule, write back
  4. Create cron job:
cron add:
  name: "time-capsule:{id}"
  schedule: kind "at", at "{openAt ISO8601}"
  sessionTarget: "isolated"
  payload: kind "agentTurn"
    message: "[TIME CAPSULE] Capsule {id} is due. Read ~/.openclaw/state/time-capsules.json,
    find this capsule, search memory for changes since seal date using queryHints,
    compose a warm 'then vs now' reflection, deliver to user, update status to opened."
    timeoutSeconds: 300
  delivery: mode "announce"
  deleteAfterRun: true
  1. Confirm with a warm message. Include capsule emoji, open date, and days remaining.

Do NOT repeat the full sealed content — it's sealed now!

Seal confirmation examples

Baby milestone:

✉️ 时光瓶已封好 这颗「时雨成长碎片」会在 2027年1月21日 准时到达 🎂 *距离开启还有 307 天 🔮*

Career decision:

✉️ 时光瓶已封好 这份决策记录会在 2026年9月20日 回来。 到时候给自己打个分——当初的直觉准不准? *距离开启还有 184 天 🔮*

Playful bet:

✉️ 时光瓶已封好(附带公证效力 ⚖️) 这个赌局会在 2027年1月21日 开奖。 *距离开启还有 307 天 🔮*

Open Flow

When a cron fires or user asks to open early:

  1. Read capsule from JSON
  2. Search memories — 3 rounds, stop when ≥3 relevant results:

- Round 1: each queryHintmemory_search({ query: hint, maxResults: 5, minScore: 0.3 }) - Round 2: each recentTopic → broader search - Round 3: generic time-range search ("recent activities", user's life events)

  1. Compose reflection (150-300 chars Chinese / 200-400 chars English):

- Quote the sealed message in blockquote - If memories found: narrate changes as a story, NOT bullet-point comparison - If few/no memories: honest "letter from the past" tone — never fabricate - End with one open question inviting user to share what's changed

  1. Adapt voice to mood:

- excited → light, smiling: "当时的你兴冲冲地说了这段话——" - anxious → warm, grounding: "那时候你挺纠结的。" - sad/grief → very gentle, more whitespace: "这是一段安静的记录。" - determined → straightforward, respectful: "当时的你说得很坚定。" - playful → fun, teasing: "让我们揭晓赌局结果——"

  1. Update capsule: status: "opened", set openedAt, save brief reflection summary
  2. Write updated JSON
  3. Deliver with action buttons: [💬 回复这颗胶囊] [📦 再封存一颗]

Open format template

🌅 一个时光瓶到期了

📅 封存于 {createdAt readable}({days}天前)

当时的你说:

> "{message}"

{reflection — 3-5 sentences, narrative style}

[💬 回复这颗胶囊]  [📦 再封存一颗]

Fallback when no memory found

这一年里的事我不太清楚——但那时候的你认真地写下了这些话, 说明那个时刻对你很重要。 现在回头看,你会想跟那时候的自己说什么?

Sensitive Content Handling

When mood is sad/grief or content mentions loss/illness/death:

  • Sealing: Don't comment on the sadness. Just: "收到了。我帮你好好收着。"
  • Opening: Extra gentle. Minimal analysis. More whitespace. No assumptions about outcomes.
  • Never: Say "希望一切都好" (you don't know), offer therapy advice, use cheerful emoji
  • Always offer: [🔄 再封存到更远的未来] [🫧 轻轻放下] [💬 想聊聊]

If user seems upset after reading:

嗯,有些记忆回来的时候会带着重量。不用急着消化。 想聊聊可以随时说,不想聊也完全 OK。

Manage Capsules

  • List: Read JSON → show sealed capsules with open dates, days remaining, first ~20 chars

- Format: 🔮 你的时光瓶 + numbered list + 共 N 个时光瓶在漂流中……

  • Delete: Require confirmation ("这颗胶囊承载了一段记忆,确定删除吗?")

→ Also cron remove the matching cronJobId

  • Modify time: Update openAt in JSON + remove old cron + create new cron
  • Open early: Gently discourage ("提前打开的话,就少了'重新遇见'的感觉。确定吗?")

but never block — user has absolute control

Pre-arrival Nudge

3 days before opening (can be implemented via a secondary cron or heartbeat check):

🔮 你有一个时光瓶将在 3 天后 开启。到时候见 ☺️

Edge Cases

  • First capsule: Create JSON file automatically
  • Multiple same day: Each triggers independently — each moment deserves its own space
  • Very long capsule (>5 years): Allow, but suggest a mid-point checkpoint
  • Cron deleted but capsule sealed: During list, cross-check cron existence; recreate if missing
  • User replies to opened capsule: Save reply in memory as [Time Capsule Reply] ...

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

72.56%
按下载量换算1,042

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

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。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