Skip to content

alchaincyf/nuwa-skill

huashu-nuwa

Verified

女娲造人:输入人名/主题/甚至只是模糊需求,自动深度调研→思维框架提炼→生成可运行的人物Skill。 两种入口:(1)明确人名→直接蒸馏 (2)模糊需求→诊断推荐→再蒸馏。 触发词:「造skill」「蒸馏XX」「女娲」「造人」「XX的思维方式」「做个XX视角」「更新XX的skill」「升级XX的skill」。 模糊需求也触发:「我想提升决策质量」「有没有一种思维方式能帮我...」「我需要一个思维顾问」。 English triggers: "distill [person]", "nuwa", "create a [person] perspective skill", "how does [person] think", "I need a thinking advisor".

scripts/download_subtitles.sh

View on GitHub
#!/bin/bash
# 从YouTube视频下载字幕
# 用法: ./download_subtitles.sh <YouTube_URL> [输出目录]
# 优先下载人工字幕,无人工字幕则下载自动生成字幕
# 语言优先级:中文 > 英文 > 其他
# 输出 SRT 或 VTT(YouTube 通常只提供 VTT),两种格式 srt_to_transcript.py 都能处理

set -e

URL="$1"
OUTPUT_DIR="${2:-.}"

if [ -z "$URL" ]; then
    echo "用法: ./download_subtitles.sh <YouTube_URL> [输出目录]"
    exit 1
fi

mkdir -p "$OUTPUT_DIR"

# 每次尝试先下载到独立的临时目录,确保只认本次新下载的文件
WORK_DIR=$(mktemp -d)
trap 'rm -rf "$WORK_DIR"' EXIT

# 用法: try_download <yt-dlp字幕参数> <语言列表>
# 下载成功则把字幕移到输出目录并打印路径,否则返回非零
try_download() {
    rm -rf "${WORK_DIR:?}"/*
    yt-dlp "$1" --sub-langs "$2" --sub-format "srt/vtt/best" --skip-download \
        -o "$WORK_DIR/%(title)s" "$URL" >/dev/null 2>&1 || true
    local file
    file=$(find "$WORK_DIR" -type f \( -name "*.srt" -o -name "*.vtt" \) | head -1)
    [ -n "$file" ] || return 1
    mv "$file" "$OUTPUT_DIR/"
    echo "$OUTPUT_DIR/$(basename "$file")"
}

echo ">>> 检查可用字幕..."
yt-dlp --list-subs --no-download "$URL" 2>/dev/null | tail -20 || true

echo ""
echo ">>> 尝试下载人工字幕(中文优先)..."

# 尝试1: 人工中文字幕
if FOUND=$(try_download --write-subs "zh-Hans,zh-Hant,zh,zh-CN,zh-TW"); then
    echo "✅ 下载成功: $FOUND"
    exit 0
fi

# 尝试2: 人工英文字幕
echo ">>> 无中文人工字幕,尝试英文..."
if FOUND=$(try_download --write-subs "en,en-US,en-GB"); then
    echo "✅ 下载成功: $FOUND"
    exit 0
fi

# 尝试3: 自动生成字幕(中文优先)
echo ">>> 无人工字幕,尝试自动生成字幕..."
if FOUND=$(try_download --write-auto-subs "zh-Hans,zh,en"); then
    echo "✅ 自动字幕下载成功: $FOUND"
    exit 0
fi

echo "❌ 未找到任何可用字幕"
exit 1
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall