跳到正文
cccskills
免费在 GitHub 公开

codexkit-meeting-summary-generator

Summarize meeting notes or transcripts into decisions, action items, owners, risks, and follow-up messages. Use after meetings when raw notes need a clean recap.

中文简介正在准备中。当前显示原文说明。

集成工具Slack
查看安装方法

包含的文件(2)

  • SKILL.md3.9 KB
  • agents/openai.yaml195 B

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Meeting Summary Generator

When to Use

  • Turning a transcript, rough notes, or chat log into a meeting recap.
  • Extracting decisions, action items, owners, due dates, and unresolved questions.
  • Preparing a follow-up email or project tracker update after a meeting.
  • Cleaning up a meeting where participants left with unclear next steps.

Procedure

Step 1 - Identify Meeting Context

Capture meeting title, date, attendees, purpose, and expected decision or outcome.

Step 2 - Extract Decisions

List only decisions that were actually made. If the notes imply a decision but do not confirm it, mark it as "proposed" or "needs confirmation."

Step 3 - Extract Actions

For each action item, identify:

  • owner
  • task
  • due date or suggested date
  • dependency
  • source quote or note if useful

Step 4 - Surface Open Questions

Separate questions that block work from questions that are informational.

Step 5 - Draft Follow-up

Write a short recap suitable for email or Slack. Keep it factual and avoid attributing commitments without evidence.

Inputs

InputRequiredFormat
Notes or transcriptYesText, bullets, transcript excerpt
Meeting purposeRecommendedOne sentence
AttendeesRecommendedNames and roles
Desired recap formatOptionalEmail, Slack, project tracker

Output

## Meeting Summary - [Meeting Name]

### Executive Recap
[3-5 sentences]

### Decisions
| Decision | Owner | Evidence / Note |
|----------|-------|-----------------|

### Action Items
| Action | Owner | Due Date | Dependency | Status |
|--------|-------|----------|------------|--------|

### Open Questions
| Question | Blocking? | Suggested Next Step |
|----------|-----------|---------------------|

### Follow-up Draft
[Ready-to-send recap]

Quality Criteria

  • Decisions are separated from discussion points.
  • Every action item has an owner or is marked "owner missing."
  • Dates are preserved exactly when provided and marked as suggested when inferred.
  • The recap is neutral, factual, and not inflated with generic language.
  • Confidential or sensitive content is summarized carefully and flagged when needed.

Verification (4C)

CheckQuestion
CorrectnessAre decisions and action items supported by the notes rather than inferred too aggressively?
CompletenessAre decisions, actions, open questions, risks, and follow-up needs all captured?
Context-fitIs the recap appropriate for the audience and meeting type?
ConsequenceCould any person be assigned work they did not actually accept?

Edge Cases

  • Transcript is messy or speaker labels are missing - Extract themes and mark ownership as uncertain.
  • No decisions were made - State that clearly and list unresolved decision points.
  • Conflicting action ownership - Flag the conflict and ask for confirmation instead of choosing silently.
  • Sensitive HR/legal content - Produce an internal summary only and recommend human review before distribution.

Examples

Prompt: "Summarize this weekly leadership meeting transcript into decisions, actions, open questions, and a Slack follow-up."

Good pattern: "Decision: Launch date remains June 12. Owner: Maya. Evidence: CEO confirmed no date change." This is better than "The team discussed launch timing."

Definition of Done

  • The output can be pasted into a recap email with minimal editing.
  • Each action has an owner or an explicit ownership gap.
  • Open questions are visible and prioritized.
  • No unsupported commitments are invented.

Changelog

  • v1.0.0 - Initial release

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Design rigorous A/B test plans with hypothesis, sample size calculation, Minimum Detectable Effect (MDE), randomization strategy, and decision rules. Includes guardrail metrics and rollout playbook. Use when planning product experiments, conversion optimization, or data-driven feature decisions.

中文简介正在准备中。当前显示原文说明。

hoavdc/CodexKit252026年10月11日 更新

Review REST and GraphQL API designs for consistency, usability, and best practices. Covers naming conventions, versioning strategy, error format, pagination, authentication patterns, and breaking change detection. Use when reviewing API specs, designing new APIs, or auditing existing endpoints.

中文简介正在准备中。当前显示原文说明。

hoavdc/CodexKit252026年10月11日 更新

Write Architecture Decision Records (ADRs) following the Michael Nygard format. Captures context, options considered, decision rationale, and consequences. Use when making technology choices, framework selections, or any architectural decision that future developers need to understand.

中文简介正在准备中。当前显示原文说明。

hoavdc/CodexKit252026年10月11日 更新

Assess organizational readiness for financial audits (internal or external). Map assertions to account balances, check evidence completeness, score readiness using a Red/Amber/Green framework, and generate a remediation timeline. Aligned with SOX, IFRS, and GAAP audit standards. Use before scheduled audits or when preparing for first-time compliance.

中文简介正在准备中。当前显示原文说明。

hoavdc/CodexKit252026年10月11日 更新

Design safe recurring Codex automations with clear prompts, outputs, schedules, and gating rules.

中文简介正在准备中。当前显示原文说明。

hoavdc/CodexKit252026年10月11日 更新

Refine Product Backlog Items to meet INVEST criteria. Write User Stories with Acceptance Criteria in Given/When/Then format, estimate with Story Points, and flag dependencies. Use before sprint planning when backlog items need grooming. Do not use to prioritize the backlog — that is the Product Owner's decision.

中文简介正在准备中。当前显示原文说明。

hoavdc/CodexKit252026年10月11日 更新

查看 hoavdc 的全部技能

报告此技能的问题