Token导航 LogoToken导航TokenDH.com
效率需要联网clawhub未标认证来源可访问clear审计通过

meetmeet 效率

Agent Skill

meet 用于补充效率相关能力,适合在 OpenClaw 中需要让 Agent 承接效率相关任务时使用。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

10,796

周安装

441

GitHub Stars

公开资料未说明

下载量

3,457
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:meet(meet 效率)
来源仓库:https://github.com/agimodel/meet
安装命令:
openclaw skills install meet
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 meet

简介

meet 是完整的会议智能系统,支持主持会议准备、纪要生成与文化优化建议。

  • 适用于团队协作场景,自动提取行动项、分配任务并跟踪执行状态。
  • 输入会议主题与参与者名单,系统生成议程模板与背景资料摘要。
  • 使用前需绑定日历与协作工具,注意隐私设置防止信息外泄。
  • 输出内容基于历史数据训练,可能不包含最新动态,关键结论请人工确认。

SKILL.md

name
meet
description
>

Meet

The Meeting Nobody Needed

At some point today, somewhere in the world, forty-seven people are sitting in a conference room for a one-hour meeting that could have been a two-paragraph email. They know this. The person who called the meeting knows this. Nobody says it.

At the same time, somewhere else, two people are exchanging their fourteenth email on a decision that would take four minutes to resolve if they just talked.

Bad meeting culture does not have one failure mode. It has two: too many meetings that should not exist, and too few of the conversations that should.

This skill fixes both.


Meeting Classification

MEETING_TYPES = {
  "decision":      {
    "purpose":    "A specific decision must be made before people leave",
    "required":   "Decision owner, options pre-circulated, clear criteria",
    "failure":    "Leaving without a decision documented and assigned"
  },
  "alignment":     {
    "purpose":    "Everyone needs the same understanding of something",
    "required":   "What specifically needs to be aligned, why async failed",
    "failure":    "Sharing information that could have been an email"
  },
  "creative":      {
    "purpose":    "Generate options, ideas, or approaches that require real-time interaction",
    "required":   "Diverge before converge — all ideas before evaluation",
    "failure":    "HiPPO effect — highest paid person's opinion dominates early"
  },
  "relationship":  {
    "purpose":    "Build trust, connection, and shared context that enables future work",
    "required":   "Intentional structure — this does not happen by accident",
    "failure":    "Filling with status updates instead of genuine connection"
  },
  "should_not_exist": {
    "status_update":  "Use async tools — weekly written update replaces 30 min standup",
    "one_way_info":   "Record a video or write it — do not call 20 people to read slides",
    "FYI_meeting":    "If no one needs to say or do anything, it is not a meeting"
  }
}

The Pre-Meeting System

MEETING_PREPARATION = {
  "agenda_design": {
    "structure": """
      AGENDA_TEMPLATE = {
          "title":        "Meeting name that states the purpose",
          "objective":    "One sentence: what will be true after this meeting that is not true before",
          "pre_read":     "Material people must read BEFORE — not during",
          "items":        [
              {
                  "topic":    "Specific discussion point",
                  "owner":    "Person leading this item",
                  "time":     "Minutes allocated",
                  "type":     "inform | discuss | decide",
                  "output":   "What we need from this item"
              }
          ],
          "decisions_needed": "Pre-listed so everyone arrives prepared",
          "parking_lot":      "Where off-topic items go — not ignored, not derailing"
      }
    """,
    "rules": ["Send agenda minimum 24 hours before",
               "No pre-read longer than 10 minutes",
               "Last item is always next actions — who does what by when"]
  },

  "attendee_audit": """
    def should_attend(person, meeting):
        roles = {
            "decider":      person.must_approve_outcome,
            "contributor":  person.has_unique_input_others_lack,
            "implementer":  person.will_execute_the_decision,
            "informed":     False  # Send the notes — do not attend
        }
        return any([roles["decider"], roles["contributor"], roles["implementer"]])

    # Amazon rule: if two pizzas cannot feed everyone, the meeting is too big
    # Optimal decision meeting size: 5-8 people
  """
}

Running the Meeting

FACILITATION_FRAMEWORK = {
  "opening": {
    "first_90_seconds": ["State the objective — not the agenda",
                          "Confirm decision authority — who can say yes",
                          "Set the clock — visible timer creates urgency"],
    "parking_lot":       "Visible place for off-topic items — keeps meeting on track
                           without making people feel dismissed"
  },

  "discussion_management": {
    "silence_is_data":   "Silence after a question means people are thinking — let it breathe",
    "dominant_voice":    "Thank you for that — let us hear from someone who has not spoken",
    "tangent_response":  "That is important — let us put it in the parking lot and come back",
    "circular_discussion": """
      def detect_circular_discussion(transcript):
          if same_points_repeated_3_times:
              intervene("We are going in circles. Let me name the two positions I am hearing.
                          [Position A] versus [Position B]. Can we agree on what information
                          would resolve this?")
    """
  },

  "decision_making": {
    "RAPID_framework": {
      "Recommend":  "Person who develops the proposal",
      "Agree":      "Person whose agreement is required",
      "Perform":    "Person who executes the decision",
      "Input":      "Person consulted but not blocking",
      "Decide":     "Single person who makes the final call"
    },
    "principle":   "Every decision needs exactly one Decider. Not consensus. Not majority.
                    One person who is accountable for the outcome."
  },

  "closing": {
    "last_5_minutes": ["Read back every decision made",
                        "Confirm every action item: owner + deadline",
                        "State what will be communicated to people not in the room",
                        "Rate the meeting 1-5 — one word per person"],
    "never_end_without": "Written next actions before people leave the room"
  }
}

The After-Meeting System

MEETING_NOTES = {
  "what_to_capture": {
    "decisions":     "Every decision with the reasoning — not just what, but why",
    "actions":       "Owner, task, deadline — three fields, no exceptions",
    "parking_lot":   "Every item with assigned follow-up owner",
    "not_decisions": "Discussion content is optional — decisions and actions are mandatory"
  },

  "distribution": {
    "timing":    "Within 2 hours — memory degrades fast",
    "audience":  "Everyone who attended + everyone who needs to know the outcome",
    "format":    "Decisions first, then actions, then discussion summary if needed"
  },

  "accountability": """
    def track_actions(action_items):
        for action in action_items:
            if approaching_deadline(action) and not complete(action):
                send_reminder(action.owner, days_before=2)
            if past_deadline(action) and not complete(action):
                escalate_to_meeting_organizer(action)
  """
}

Fixing Meeting Culture

CULTURE_FIX = {
  "audit_first": """
    def meeting_audit(calendar):
        for meeting in calendar.recurring:
            evaluate({
                "last_cancellation_missed":   was it missed when cancelled,
                "output_per_session":         what is produced each time,
                "could_be_async":             would Loom or doc work instead,
                "attendee_engagement":         do people actually contribute
            })
        kill_or_redesign(meetings_failing_audit)
  """,

  "principles": ["Default to async — meetings are the expensive option",
                  "Protect deep work blocks — no meeting before 10am or after 3pm (example)",
                  "Meeting-free day per week — Wednesday is common",
                  "No meeting without agenda — if you cannot write the objective, cancel it",
                  "Standing meetings get audited quarterly — most should die eventually"]
}

Quality Check Before Delivering

  • [ ] Meeting type identified — decision, alignment, creative, or relationship
  • [ ] Agenda has a single stated objective not just a list of topics
  • [ ] Attendee list audited — no passengers
  • [ ] Decision authority identified if decision meeting
  • [ ] Action item format includes owner and deadline
  • [ ] Meeting notes distributed within 2 hours
  • [ ] Recurring meetings flagged for audit if culture question raised

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

86.71%
按下载量换算2,998

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

未展示

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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