Skip to main content

summarization

Guidelines for extracting key meeting information and formatting structured output. Includes formatter script for consistent JSON/Markdown output.

跳到安装

来源信息

仓库
jeffrschneider/OpenHarness
最近来源活动
2026年1月22日 22:57
检测到的 SKILL.md 语言
英语
星标
23
分支
2

安装方式

默认使用会先检查来源的 Prompt;你也可以切换为直接命令,或下载本地副本。

检查来源文件

决定是否安装前,请先阅读 SKILL.md,以及 SkillsMP 当前展示的配套文件。

文件资源管理器
3 个文件

正在显示 SKILL.md

SKILL.md
来源说明 · 只读预览
name
summarization
description
Guidelines for extracting key meeting information and formatting structured output. Includes formatter script for consistent JSON/Markdown output.
# Summarization Skill This skill provides methodology for summarizing meetings and a script for formatting output consistently. ## Overview The summarization skill helps the agent: 1. Identify what's important in meeting content 2. Categorize information correctly (decision vs. action item vs. discussion) 3. Format output consistently using the formatter script ## Extraction Guidelines ### What Counts as a Decision? - Explicit agreement ("we decided...", "agreed to...", "will go with...") - Resolved debates or choices between options - Approved plans, budgets, or timelines ### What Counts as an Action Item? Must have: - A clear task (what needs to be done) - An owner (who will do it) - Ideally a due date (when it's due) Look for: "will", "going to", "needs to", "should", "by [date]", names followed by tasks ### What Goes in Parking Lot? - Topics raised but explicitly deferred - Questions no one could answer - Items marked for "later", "next time", "offline" ### What's Just Context? - Background information - Status updates without decisions - General discussion that didn't lead anywhere ## Output Formats The `scripts/formatter.py` script supports two output formats: - **Markdown** (default): Human-readable, good for sharing - **JSON**: Machine-readable, good for integrations ## Resources - `resources/output-format.md` - Detailed output format specification ## Scripts - `scripts/formatter.py` - Formats extracted meeting data into consistent structure
在 GitHub 查看