使い方
- インストールコマンドをコピーし、自分の環境へ導入する前にソースを確認します。
- 必要なツール、認証情報、対応環境を原文で確認します。
- 入力と成果物を指定し、出力を資料と照合します。
npx skills add https://github.com/mohammedsadeghi/youtube-agent-skill --skill yt-retention開始前の確認
- 導入にはモデル契約、API枠、ファイルへのアクセス許可は含まれません。
- スター数や導入数は活動の指標であり、独立した品質・安全性試験ではありません。
見出し、例、コード、表、リンク、参照画像を含む原文を表示しています。
yt-retention
The retention graph shows where viewers keep watching and where they leave. Almost nobody exports it.
python3 "$SKILL_DIR/retention.py" retention.csv --duration 600
python3 "$SKILL_DIR/retention.py" retention.csv --transcript transcript.srtBefore running the helper, set SKILL_DIR to the full path to the directory containing this SKILL.md. Claude Code provides that path as CLAUDE_SKILL_DIR; other agents should resolve it from the loaded skill file. Do not assume the user's working directory is the skill directory.
Getting the file: Studio -> a video -> Analytics -> Engagement -> the audience-retention chart -> the download icon -> "Audience retention".
Three different problems
- HOOK LEAK - what is lost in the first 30 seconds. Under 25% is healthy. This is always the
first thing to fix and it is always the first fifteen seconds of script, never the edit.
- CLIFFS - single steep drops. A cliff is a moment: a topic change with no signposting, a
sponsor read, a long setup. With --transcript the tool prints what was being said there, which is what makes the report actionable instead of interesting.
- SLIDE - the steady bleed across the middle. A flat slide is pacing. The fix is cutting, not
rewriting.
What to hand back
Name the single biggest leak and one change for it. Not a list of five. Then, only if asked, the rest. If the hook leak is healthy and the slide is flat, say the video is fine and review packaging with yt-package if that skill is installed.
The gate
Nothing here publishes. This skill writes and you publish. Every output ends in a block the user copies, and the last line of every run is the question: ship it, or change it?