看懂视频(watch-anything)
SkillMediaUnderstands videos and rich media: X posts (including tweet videos) / YouTube / Bilibili / video links without subtitles → subtitle-first or local/cloud speech transcription → Chinese card-style breakdown (card: 3-sentence summary + 3-6 key points + bilingual standout quotes; full transcript: narrat
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the 看懂视频(watch-anything) skill
What this skill tells your AI
The instructions your AI receives, as published by heyuxuan0209/watch-anything in SKILL.md and read by ahel’s review.
按本目录 PLAYBOOK.md 执行,它是唯一的行为规范(路由表、转写降级链、卡片格式、
诚实守则、降级话术都在里面)。本文件只是 Claude Code 的触发入口。
要点提醒:
- 先读
PLAYBOOK.md,按链接类型走路由:X 推文先scripts/fetch-x.mjs拿正文+媒体清单, 带视频再scripts/transcribe-video.mjs转写;YouTube/B站 直接transcribe-video.mjs(字幕优先,无字幕才转写)。脚本零 npm 依赖、Node 18+、JSON 进出。 - 转写降级链:视频字幕 > Groq 云 whisper(需
GROQ_API_KEY)> 本地 faster-whisper。 缺哪级降哪级,长视频转写前告知用户「需等几分钟」。 - 解读模板在
templates/,默认card.md(卡片格式);用户一句话可覆盖, 自定义模板 = 目录里的新 .md 文件。 输出终点在integrations/(可选):环境里有FEISHU_WEBHOOK/FEISHU_APP_ID时, 卡片出完可问一句要不要推飞书,跑scripts/push-feishu.mjs;没配就当它不存在,别催用户配。 - 诚实守则优先级最高:抓不到的推文、转不出的音频、未解析的画面,如实降级并显式声明, 绝不脑补视频/图片里的内容。
Signals
- GitHub stars
- 24
- Forks
- 1
- Last commit
- Aug 2026
Advanced
- Catalog kind
- skill
- Gateway key
watch-anything- Source
- github.com/heyuxuan0209/watch-anything