Menu
Download slide images from a SpeakerDeck presentation. Use when the user provides a SpeakerDeck URL and wants to read, summarize, or convert the slides.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
トレンド収集
gog CLIでGmailの過去90日を読み取り専用監査し、頻繁かつ低価値なメールマガジンや反復通知だけを最大5件の解除・配信頻度変更・filter候補として提案する。受信箱のノイズ低減、newsletter整理、定期的なメール衛生レビューで使う。低品質でも低頻度のメールは候補にせず、解除、filter作成、削除、既読化は実行しない。
GmailとSlackの直近メッセージから、根拠付きの「今日やること」を最大3件に絞り、ADHDフレンドリーな日次アクションプランとしてlifeリポジトリへ保存する。朝や昼の受信箱トリアージ、返信・確認・フォローアップの計画、Hermes Agentの定期実行で使う。メール送信、Slack投稿、既読化、タスク登録、予定登録は行わない。
Conduct large-scale, resumable research over public X/Twitter posts with Bird from a user-supplied research specification. Use when collecting hundreds to tens of thousands of posts, enriching public profile signals, applying an arbitrary evidence-based classification rubric, and generating a source-linked dashboard with original-post detail views.
X/Twitter command-line tool for reading, searching, posting, and interacting. Use when the user wants to read, search, post, or interact with X/Twitter through the bird CLI.
Grokを使ってX（Twitter）上のリアルタイム情報を検索する。ユーザーの反応、最新トレンド、議論の調査に使用。Chrome MCP環境でのみ動作。Triggers on: 'Xでの反応', 'ツイッターで', 'SNSの反応', '最新情報', 'リアルタイム', 'トレンド', '話題', 'バズ', '評判', '口コミ', 'みんなどう言ってる', 'みんなの反応', '反応を調べ', '〜の反応', '〜への反応', '世間の声', '感想を調べ', 'grokで', 'grokを使って', '〜の使い方', '〜のtips', 'tips', '便利な使い方', 'おすすめ設定'
| name | speakerdeck |
| description | Download slide images from a SpeakerDeck presentation. Use when the user provides a SpeakerDeck URL and wants to read, summarize, or convert the slides. |
| user-invocable | true |
Download all slide images from a SpeakerDeck presentation URL. Zero dependencies — uses only built-in fetch and fs APIs. The downloaded images can be read by Claude Vision or converted with markitdown.
nix-shell -p bun --run "bun {baseDir}/scripts/speakerdeck.ts <speakerdeck-url> [-o output-dir]"
| Arg | Description |
|---|---|
<url> | SpeakerDeck presentation URL (required) |
-o, --output <dir> | Output directory (default: $TMPDIR/speakerdeck-<id>/) |
{
"title": "Presentation Title",
"presentationId": "abc123def456",
"slideCount": 30,
"outputDir": "/tmp/speakerdeck-abc123def456",
"files": ["/tmp/speakerdeck-abc123def456/slide_000.jpg", "..."]
}
result=$(nix-shell -p bun --run "bun {baseDir}/scripts/speakerdeck.ts 'https://speakerdeck.com/user/talk'")
# Then use the file paths from the JSON output with Read tool
result=$(nix-shell -p bun --run "bun {baseDir}/scripts/speakerdeck.ts 'https://speakerdeck.com/user/talk'")
dir=$(echo "$result" | jq -r '.outputDir')
for f in "$dir"/*.jpg; do markitdown "$f"; done
files.speakerdeck.com) to determine slide countnix-shell -p bun