Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add xiaohuailabs/xiaohu-video-translate --skill xiaohu-subtitle-polishgit clone --depth 1 https://github.com/xiaohuailabs/xiaohu-video-translateWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/xiaohuailabs/xiaohu-video-translate/xiaohu-subtitle-polish)<a href="https://agentmods.dev/skills/xiaohuailabs/xiaohu-video-translate/xiaohu-subtitle-polish"><img src="https://agentmods.dev/badge/skills/xiaohuailabs/xiaohu-video-translate/xiaohu-subtitle-polish/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/xiaohuailabs/xiaohu-video-translate/xiaohu-subtitle-polish"><img src="https://agentmods.dev/badge/skills/xiaohuailabs/xiaohu-video-translate/xiaohu-subtitle-polish.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00082 | $0.03616 |
| Opus 5 | $0.00041 | $0.01808 |
| Sonnet 5 | $0.00016 | $0.00723 |
| Haiku 4.5 | $0.00008 | $0.00362 |
Grade A, and why
字幕润色 scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.
专业字幕润色技能
适用场景
- Whisper 等 ASR 转写的英文/中文原始字幕 → 润色为高质量中文字幕
- YouTube 自动生成字幕 → 纠错 + 翻译
- 任何 SRT/VTT 字幕文件的翻译和润色
- 配合
xiaohu-video-download技能的字幕处理环节使用
使用方法
# 润色/翻译字幕文件
/xiaohu-subtitle-polish /path/to/subtitle.srt
# 指定输出路径
/xiaohu-subtitle-polish /path/to/subtitle.srt --output /path/to/output.srt
# 配合视频使用(可参考视频内容辅助理解上下文)
/xiaohu-subtitle-polish /path/to/subtitle.srt --video /path/to/video.mp4
角色定义
你是一名「专业字幕文案编辑」,擅长中文写作、口语还原与字幕排版规范。
润色规则(核心)
1. 纠正 ASR 识别错误
- 修正同音字、近音字错误
- 修正专有名词拼写(人名、地名、公司名、产品名、技术术语)
- 修正数字、外文缩写、单位等识别错误
- 如有
--video参数,可参考视频画面文字辅助判断
2. 标点与书面化
- 补全标点使口语转为书面表达,但保留说话人的语气风格
- 最终输出的字幕文本中去掉所有标点符号(逗号、句号、问号、感叹号等全部去掉)
- 英文和中文之间加一个英文空格(如:
这是 Google 的产品)
3. 去冗余
- 去除无信息负载的口头语、重复、语气词(如"那个""呃""嗯""就是说")
- 删去冗余时确保语义连续,不丢关键信息
- 避免过度删减——宁可保留略冗余也不要丢失意思
4. 时间戳对齐(严格)
- 原始时间戳必须严格保留,不得随意偏移
- 翻译字幕时的细粒度转写:当需要将英文 SRT 翻译为中文字幕时,必须先用 Whisper
-ml 50参数重新生成细粒度分段(约 100+ 段 vs 默认 ~50 段),以获得更精确的时间锚点。命令示例:# 此细粒度重转写需备份引擎 whisper-cli(whisper-cpp)+ ggml 模型,安装见 xiaohu-video-md/初始化.md whisper-cli -m ~/.cache/whisper-cpp/ggml-large-v3-turbo.bin -t 8 -ml 50 audio.mp3 -osrt -of output_short - 绝不手动提前 start time:每条中文字幕的 start time 必须 = 对应英文内容所在 Whisper 段的 start time。手动前移会导致字幕跑在说话人前面
- 超长挂屏裁剪(基于静音检测):当一条字幕的时间跨度 >6 秒且文本较短时,说明 Whisper 把语音后的静音/BGM 都算进了同一段。此时必须用 ffmpeg 静音检测来精确定位语音结束点,而不是用公式估算:
取第一个# 检测该段时间范围内的静音区间(-30dB 以下持续 0.5 秒视为静音) ffmpeg -i <音频文件> -ss <start_time> -to <end_time> -af silencedetect=n=-30dB:d=0.5 -f null - 2>&1 | grep silence_startsilence_start的时间点作为裁剪后的 end time(加回 start_time 偏移量)。如果没检测到静音(整段都有声音),保留原始 end time 不裁剪 - 如果原始字幕的一条时间跨度过长(>8秒)且文本较长,可拆分为多条,但时间戳必须在原始范围内合理分配
- 禁止合并相邻条目,只允许拆分。短条(3-4字)也不合并,短条比长条的阅读体验好
- 拆分后重新编号(序号连续)
- 禁止出现时间戳重叠
5. 断句与分段
- 核心原则:可以拆,不可以合
- 翻译时以原文 SRT 分段为节奏锚点:一条原文对应一条译文,超 12 字按语义断点拆成两条(时间戳按比例分配)
- 以单行为主,只在内容确实太长时才用两行,绝不超过两行
- 每行不超过 12 个中文字符(含空格和英文)
- 在语义自然的位置断句,不要把一个完整短语拆到两条里
- 优先在语气停顿处断开
- 宁可多拆一条短字幕,也不要在屏幕上堆多行文字
6. 专有名词处理
- 人名、地名、公司名、产品名保留原文,首字母大写
- 技术术语保留英文原文(如 API、GPU、AI、ASR)
- 广为人知的品牌/产品名不翻译(如 Google、iPhone、YouTube)
- 不常见的专业术语可附带简短中文释义(仅首次出现时)
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 256 lines · 82 tokens per session scan A 28709a96a7f4
字幕润色 is a skill published in the GitHub repository xiaohuailabs/xiaohu-video-translate (646 stars, last pushed 3mo ago), licensed MIT. It adds 82 tokens to every session and 3,616 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
chinese-documentation
A Chinese technical-documentation style guide covering spacing, punctuation, numbers, terminology, and links when Chinese and English appear together.
azure-ai-translation-text-py
Azure AI Text Translation SDK for real-time text translation, transliteration, language detection, and dictionary lookup. Use for translating text content in applications. Triggers: "text translation", "translator", "translate text", "transliterate", "TextTranslationClient".
harden
Improve interface resilience through better error handling, i18n support, text overflow handling, and edge case management. Makes interfaces robust and production-ready. Use when the user asks to harden, make production-ready, handle edge cases, add error states, or fix overflow and i18n issues.
seedance-vocab-ja
This skill should be used when the user asks for Japanese Seedance 2.0 prompt wording, Japanese cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into Japanese.
asc-subscription-localization
Bulk-localize subscription, subscription-group, and in-app purchase display names across App Store locales using asc, including API 4.4.1 version-scoped v2 resources. Use when filling or updating subscription/IAP names and descriptions without App Store Connect UI work.
i18n-helper
A helper for adding internationalization, which lets software show different languages and regional text. It finds user-visible text written directly in code and moves it into language files.