Skip to content
Back to skills

Summarize Meetings Into Decisions

ASecurity

Extract decisions, rationale, owners, and unresolved questions from a meeting transcript.

  • 2 stars
  • 0 votes
  • 0 copies
  • 0 views
  • Added October 22, 2026
ai-agentsgoexpress

Works with

  • cursor

Security analysis

A100/100

Scanned October 22, 2026

npx -y skills add alivirgo/Major-AI-Skills --skill summarize-meetings-into-decisions --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Summarize Meetings Into Decisions?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Summarize Meetings Into Decisions
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/alivirgo-summarize-meetings-into-decisions/badge)](https://www.skillsdirectory.com/skills/alivirgo-summarize-meetings-into-decisions)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: summarize-meetings-into-decisions
description: "Extract decisions, rationale, owners, and unresolved questions from a meeting transcript."
category: common-sense
risk: safe
source: self
source_type: self
date_added: "2026-08-26"
tags: ["meeting-summaries", "decision-records", "executive-brief", "transcript-parsing", "productivity", "prompt-engineering"]
tools: ["claude", "cursor", "gemini", "codex", "chatgpt"]
---

# Summarize Meetings into Decisions (Decision Log Extraction) (AI Skill)

## Overview
When asked to summarize a 45-minute meeting transcript, standard AI outputs create a chronological play-by-play (*"First, Sarah presented the Q3 slide deck. Then Mark asked about pricing. Then Sarah replied that pricing was under review..."*). 

This chronological narrative is nearly useless for executives and team members who weren't in the room. What matters are **the final decisions agreed upon, the rationale behind them, and who is accountable for execution**.

The **Decision Log Extraction Protocol** filters out conversational chatter and formats the meeting into an **Architecture / Executive Decision Record (ADR)**.

---

## Chronological Narrative vs. Decision Record

```
┌─────────────────────────────────────────────────────────────┐
│                 Transcript Processing Models                │
│                                                             │
│  Chronological Narrative (Low Value):                       │
│  • 4 pages of "Bob said this, then Alice disagreed..."      │
│  • High reading fatigue, impossible to find final agreements│
│                                                             │
│  Decision Record (High Value):                              │
│  • [DECISION 1]: What was decided & why                     │
│  • [OPTIONS REJECTED]: What was eliminated & why            │
│  • [ACTION ASSIGNMENTS]: [Owner] + [Task] + [Deadline]      │
└─────────────────────────────────────────────────────────────┘
```

---

## Master Decision Record Prompt Templates

### Pattern 1: The Executive Decision Record (ADR)
Use for product, strategy, and leadership meetings:

```markdown
Analyze the attached meeting transcript: [ATTACH TRANSCRIPT / PASTE TEXT].

Do NOT write a chronological timeline of who spoke.
Extract a structured **Executive Decision Record**:

### 🎯 1. Key Decisions Made
For each decision, format as:
- **Decision**: [Clear 1-sentence statement of what was approved]
- **Rationale**: [Why this option won over alternatives]
- **Trade-off / Cost**: [What we are giving up or spending]

### 🚫 2. Options Explicitly Rejected
- [Option Name]: [Why the team decided NOT to proceed with this path]

### ⚡ 3. Accountable Action Items
| Task | Owner | Due Date | Definition of Done |
```

---

## Real-World Case Study

### Scenario: Product Team Pricing Sync

#### Chronological AI Output (Rambling)
> *"The meeting began at 2 PM. Dave suggested increasing enterprise pricing by 20%. Lisa expressed concern that existing customers might churn. Mark mentioned that our churn rate is currently 1.8%. After some back-and-forth discussion regarding grandfathering policies, the team decided to increase new customer pricing..."*

#### Decision Record Output (Instant Clarity)

### 🎯 1. Key Decisions Made
- **Decision**: Increase Enterprise Tier pricing by **20% (from $500/mo to $600/mo)** for all new signups starting September 1st.
- **Rationale**: Competitor benchmarks show we are 35% underpriced; low churn rate (1.8%) demonstrates strong price tolerance.
- **Grandfathering Policy**: All existing customers remain on legacy $500/mo pricing for 12 months.

### 🚫 2. Options Explicitly Rejected
- **Immediate Price Hike for Existing Customers**: Rejected due to risk of relationship damage with long-term champion accounts.

### ⚡ 3. Accountable Action Items
| Task | Owner | Due Date | Definition of Done |
| :--- | :--- | :--- | :--- |
| **Update Stripe Pricing Table** | Dave | Aug 25 | New $600/mo pricing tier live on staging. |
| **Draft Grandfathering FAQ Email** | Lisa | Aug 20 | Customer update email reviewed and approved. |

---

## Summary Best Practices
- **Ban "Who said what"**: Explicitly prompt *"Focus on team consensus and final decisions, not individual speech chronologies."*
- **Always record rejected options**: Capturing *why* an alternative was killed prevents the team from debating the exact same idea again 3 months later.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.