Token导航 LogoToken导航TokenDH.com
研究检索敏感数据clawhub未标认证来源可访问clear审计提醒

yt-summary总结

Agent Skill

yt-summary 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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

18,440

周安装

784

GitHub Stars

公开资料未说明

下载量

6,460
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:yt-summary(总结)
来源仓库:https://github.com/chapati23/yt-summary
安装命令:
openclaw skills install yt-summary
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 yt-summary

简介

快速总结 YouTube 视频内容,支持自定义提示词提取特定信息。

  • 适用于学习笔记整理、会议记录摘要与视频内容速览场景。
  • 粘贴 URL 并附加说明即可生成结构化摘要,提升信息获取效率。
  • 依赖视频字幕或转录文本,对无字幕视频效果有限。yt-summary 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 建议核实摘要准确性,尤其涉及专业或敏感话题时交叉验证。

SKILL.md

name
youtube-summary
version
1.3.2
author
giskard
description
Summarize any YouTube video by dropping the link in chat. Supports custom prompts — paste the URL followed by your instructions (e.g. 'focus on the technical details'). Triggers on YouTube URLs.
tags
[youtube, video, summary, transcript]
license
MIT
homepage
https://github.com/chapati23
metadata
openclaw
emoji
📺
requires
bins
[python3]
env
[TRANSCRIPT_API_KEY]
primaryEnv
TRANSCRIPT_API_KEY

YouTube Summary Skill

Summarize YouTube videos by extracting transcripts via TranscriptAPI.com and generating structured summaries.

Setup

Prerequisites

  • Python 3.10+
  • A TranscriptAPI.com account ($5/mo for 1,000 transcripts)
  • Optional: pass (Unix password manager) for secure key storage

Installation

  1. Sign up at transcriptapi.com and get your API key
  2. Provide the API key via one of these methods:

- Environment variable (simplest): export TRANSCRIPT_API_KEY="your-key-here" - pass password store (most secure): pass insert transcriptapi/api-key

  1. Install Python dependencies:
   pip install -r skills/youtube-summary/requirements.txt

Detection

Trigger on messages containing YouTube URLs matching any of:

  • youtube.com/watch?v=ID
  • youtu.be/ID
  • youtube.com/shorts/ID
  • m.youtube.com/watch?v=ID
  • youtube.com/live/ID

⚠️ Critical Rules

  • NEVER use web_search as a fallback. If transcript extraction fails, report the error and stop.
  • NEVER fabricate transcript content. Only summarize what the extraction script returns.
  • Always run the extraction script. Do not skip it, even for well-known videos.

Workflow

Step 1: Extract transcript

If using pass:

_yt_key_file=$(mktemp) && pass transcriptapi/api-key > "$_yt_key_file" && python3 skills/youtube-summary/scripts/extract.py "YOUTUBE_URL_OR_ID" --api-key-file "$_yt_key_file"; rm -f "$_yt_key_file"

If using env var:

python3 skills/youtube-summary/scripts/extract.py "YOUTUBE_URL_OR_ID"

(Reads TRANSCRIPT_API_KEY from the environment automatically.)

Security note: The pass + temp file approach avoids exposing the key in ps output or shell history. The env var approach is simpler but the key is visible in the process environment.

Parse stdout:

  • PROGRESS: lines → relay to user as status updates (optional)
  • ERROR: lines → relay error to user, stop
  • RESULT: line → parse the JSON after RESULT: — contains: header, transcript, language, tokens, title, channel, duration_str

Step 2: Summarize the transcript

Use the extracted transcript to generate a summary. The summary language must match the transcript language (from the language field).

If tokens < 50000 — single-pass: summarize the full transcript in one request.

If tokens ≥ 50000 — tell the user it's a long video and summarize the first ~40K tokens with a note that it was truncated.

Default summary format (use when no custom prompt given):

{header}

**TL;DR:** 2-3 sentence summary.

**Key Points:**
• Point one
• Point two
• (3-7 total)

**Notable Quotes:** (only if genuinely quotable lines exist)
> "Quote here"

Custom prompt — if the user included text alongside the URL, append it as additional instructions for the summary.

Step 3: Reply

  • Keep output under 4000 characters for Telegram
  • If the summary would exceed 4000 chars, send the TL;DR first, then the rest as a follow-up
  • Always include the header line from the extraction result

Error Handling

  • ERROR: API_ERROR: Invalid API key → "TranscriptAPI key is invalid. Check pass transcriptapi/api-key."
  • ERROR: No transcript available → "This video doesn't have captions available."
  • ERROR: Video not found → "Couldn't find that video — double-check the URL."
  • Any other ERROR: → relay the message as-is. Do NOT fall back to web_search.

Why TranscriptAPI?

YouTube aggressively blocks datacenter/IPv6 ranges from accessing transcripts. Most cloud VPS (Hetzner, DigitalOcean, AWS, etc.) are blocked — direct transcript fetching fails for most videos when running from a server.

TranscriptAPI.com proxies requests through residential IPs, bypassing these blocks reliably. The $5/mo plan covers 1,000 transcript fetches.

💡 Tip: Add instructions after the URL to customize the summary (e.g. "focus on the technical details").

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

71.41%
按下载量换算4,613

安全审计

VirusTotal

可疑

ClawScan

通过

Static analysis

未展示

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。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