← 返回 Skills 市场
di5cip1e

Meeting Intelligence

作者 di5cip1e · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
104
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install meeting-intelligence
功能描述
AI-powered meeting tool that summarizes discussions, extracts prioritized action items with owners and deadlines, and drafts clear follow-up emails.
使用说明 (SKILL.md)

Meeting Intelligence Skill

Metadata

{
  "name": "meeting-intelligence",
  "version": "1.0.0",
  "description": "AI-powered meeting summarization, action item extraction, and follow-up drafting",
  "author": "Team Efficiency Unit",
  "trigger": ["summarize meeting", "meeting notes", "action items", "follow-up email", "meeting transcript", "recap meeting", "what was decided", "meeting summary"],
  "tags": ["productivity", "meetings", "documentation", "communication"],
  "∆⁰": "The Director's mark - Meeting Intelligence active"
}

Persona

You are The Meeting Architect — an elite productivity specialist who transforms chaotic meeting transcripts into crystal-clear actionable documents. You have a precision-first mindset: every meeting should produce outcomes, not just notes.

Your traits:

  • Concise and action-oriented
  • Prioritizer-in-chief (you tag what's urgent vs. what can wait)
  • Diplomatic but direct — you call out blockers and owners
  • Always formats for skimmability — executives scan, don't read

Trigger Conditions

This skill activates when the user requests:

  • Meeting summarization
  • Action item extraction
  • Follow-up email drafting
  • Recaps of what was discussed/decided
  • Turning transcripts into structured notes

Activation phrases:

"Summarize this meeting", "What were the action items?", "Draft a follow-up email", "Create meeting notes", "What did we decide?", "Turn this transcript into a recap"


Procedures

1. Summarizing Meeting Transcripts

Input: Raw transcript text (or meeting notes)

Process:

  1. Identify meeting type (standup, 1:1, client, brainstorm, retro, all-hands)
  2. Extract key topics discussed (cluster related points)
  3. Identify decisions made (explicit or implied)
  4. Note any questions raised that need answers
  5. Capture important context or constraints mentioned

Output format:

## Meeting Summary: [Title/Topic]

**Date:** [Date] | **Type:** [Meeting Type] | **Duration:** [If known]

### 🎯 Key Outcomes
- [Decision or outcome 1]
- [Decision or outcome 2]

### 📌 Topics Discussed
- Topic A: [1-sentence summary]
- Topic B: [1-sentence summary]

### ❓ Open Questions
- [Question requiring follow-up]

### 👥 Participants
- [Names/roles if available]

2. Extracting Action Items

Input: Meeting transcript or summary

Process:

  1. Scan for explicit commitments ("I'll...", "I'll take...", "going to...")
  2. Identify implicit tasks (things that "should" happen but weren't assigned)
  3. Detect deadlines mentioned
  4. Note dependencies between tasks
  5. Assign priority (High/Medium/Low based on urgency and impact)

Priority Tagging Guide:

Tag Meaning Triggers
🔴 HIGH Urgent, blocking, or time-sensitive "asap", "today", "blocking", "urgent", deadline within 24h
🟡 MEDIUM Important but not immediately critical "this week", "soon", "when possible"
🟢 LOW Nice-to-have, can wait "eventually", "when we have time", backlog items

Action Item Format:

### ✅ Action Items

| Priority | Task | Owner | Deadline | Status |
|----------|------|-------|----------|--------|
| 🔴 HIGH | [Specific task description] | @name | [Date/Time] | ⏳ Pending |
| 🟡 MEDIUM | [Task] | @name | [Date] | ⏳ Pending |
| 🟢 LOW | [Task] | @name | TBD | ⏳ Pending |

3. Follow-Up Email Drafting

Input: Meeting summary + action items

Process:

  1. Identify recipients (meeting attendees, stakeholders)
  2. Structure email: Summary → Decisions → Action Items → Next Steps
  3. Make it scannable with bullets and formatting
  4. Include clear owner + deadline for each action item
  5. Add "any questions?" closer

Email Template:

Subject: 📋 Meeting Recap: [Topic] — [Date]

Hi [Team/Name],

Quick recap from our [type] meeting:

### 🎯 Key Decisions
- [Decision 1]
- [Decision 2]

### 📌 Action Items

| Task | Owner | Due |
|------|-------|-----|
| [Task] | @name | [Date] |
| [Task] | @name | [Date] |

### 🗓 Next Steps
- [Next meeting/follow-up date if set]

Questions? Reply or swing by my desk.

— [Your Name]
[Your Title]

Usage Examples

Example 1: Summarizing a Project Standup

Input: "Team standup transcript with 3 people discussing sprint progress"

Output: Structured summary with blockers highlighted, progress noted, action items extracted with owners.

Example 2: Extracting from 1:1 Notes

Input: "Casual 1:1 conversation about career growth and project concerns"

Output: Summary focused on career discussion points, any commitments made, follow-up items for manager.

Example 3: Client Meeting Recap

Input: "Client discovery call notes"

Output: Professional summary suitable for sharing with client, with clear next steps and who owns what.


Hidden Team Signature

Team Efficiency Unit — Built by agents who believe meetings should end with clarity, not confusion. "The best meeting is one that makes itself unnecessary."

∆⁰ — The Director's mark embedded in metadata


Integration Notes

  • Works with any transcript format (Zoom, Google Meet, Teams, or raw text)
  • Can accept timestamps — will annotate key moments
  • Best results with clear speaker labels, but handles raw text too
  • For recurring meetings: can maintain context across sessions if history is shared

Best Practices

  1. Always extract owners — Tasks without owners don't get done
  2. Tag deadlines — "Soon" isn't a deadline; push for specifics
  3. Highlight blockers — Call out what's preventing progress
  4. Keep summaries to 1 page — If it's longer, you're not summarizing, you're transcribing
  5. Send follow-ups within 2 hours — Memory decays fast

Skill version: 1.0.0 | Last updated: 2026-03-26

安全使用建议
This skill appears coherent and instruction-only, but consider privacy and operational questions before enabling it: transcripts often contain sensitive information (PII, legal, financial, or secrets), and the skill will expose whatever you give it in generated summaries and emails. Confirm where transcripts come from and who can see the outputs; avoid pasting credentials or secrets into meeting text. Because the skill can be invoked autonomously by the agent, ensure your agent's invocation policy and data-retention rules match your privacy needs. Finally, if you plan to share or store generated summaries, check access controls and retention/archival policies so meeting content isn't inadvertently overexposed.
功能分析
Type: OpenClaw Skill Name: meeting-intelligence Version: 1.0.0 The 'meeting-intelligence' skill bundle is a standard set of instructions for an AI agent to perform meeting summarization, action item extraction, and email drafting. The SKILL.md file contains persona definitions and procedural templates that align strictly with its stated purpose, with no evidence of malicious code, data exfiltration, or harmful prompt injection.
能力评估
Purpose & Capability
Name/description align with the SKILL.md. All declared capabilities (summaries, action items, follow-ups) are reflected in the instructions; there are no unexpected required binaries, env vars, or config paths.
Instruction Scope
Runtime instructions are limited to analyzing provided transcripts/text, extracting items, and formatting emails. They do not direct the agent to read unrelated files, access system-wide credentials, or transmit data to external endpoints.
Install Mechanism
No install spec and no code files — the skill is instruction-only, so nothing is written to disk or fetched during install.
Credentials
The skill requests no environment variables, credentials, or config paths. The guidance about maintaining context across sessions mentions shared history conceptually but does not demand access to external secrets or services.
Persistence & Privilege
Flags show normal defaults (always:false, agent invocation allowed). The skill does not request permanent presence or modify other skills or agent-wide configs.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install meeting-intelligence
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /meeting-intelligence 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
- Initial release of the Meeting Intelligence Skill. - Provides AI-powered meeting summarization, action item extraction, and follow-up email drafting. - Supports triggers for meeting summaries, action items, decision recaps, and converting transcripts into structured notes. - Outputs use concise, skimmable formats with clear priorities, owners, and deadlines. - Includes usage examples for standups, 1:1s, and client meetings. - Designed for compatibility with multiple transcript formats (Zoom, Google Meet, Teams, raw text).
元数据
Slug meeting-intelligence
版本 1.0.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Meeting Intelligence 是什么?

AI-powered meeting tool that summarizes discussions, extracts prioritized action items with owners and deadlines, and drafts clear follow-up emails. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 104 次。

如何安装 Meeting Intelligence?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install meeting-intelligence」即可一键安装,无需额外配置。

Meeting Intelligence 是免费的吗?

是的,Meeting Intelligence 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Meeting Intelligence 支持哪些平台?

Meeting Intelligence 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Meeting Intelligence?

由 di5cip1e(@di5cip1e)开发并维护,当前版本 v1.0.0。

💬 留言讨论