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 renky1025/agent-skills --skill video-dubbinggit clone --depth 1 https://github.com/renky1025/agent-skillsWrote 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/renky1025/agent-skills/video-dubbing)<a href="https://agentmods.dev/skills/renky1025/agent-skills/video-dubbing"><img src="https://agentmods.dev/badge/skills/renky1025/agent-skills/video-dubbing/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/renky1025/agent-skills/video-dubbing"><img src="https://agentmods.dev/badge/skills/renky1025/agent-skills/video-dubbing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 142 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 140 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00227 | $0.02693 |
| Opus 5.5 | $0.00091 | $0.01077 |
| Sonnet 5.5 | $0.00045 | $0.00539 |
| Haiku 4.5 | $0.00023 | $0.00269 |
Grade A, and why
video-dubbing 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 1mo 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
video-dubbing: 视频翻译配音与字幕同步
Purpose
把一段视频整体本地化为另一种语言:识别原语音、翻译、用 TTS 生成目标语言配音,在保持原始 时间戳的前提下合成音轨并把译文硬字幕烧进画面。产出是可直接发布的单文件视频,配音与字幕 天然同步,不需要二次对轴。
When to Use
- 用户要求把视频翻译成另一种语言,并要配音(不只是字幕)。
- 需要给视频加目标语言配音 + 同步字幕(双语或纯译文)。
- 触发词:视频翻译、视频配音、翻译字幕、生成外语配音、双语字幕、dub a video、translate a video、 foreign-language voiceover。
- 任何 ASR -> 翻译 -> TTS -> 视频合成的工作流。
When NOT to Use
- 只需转录、不需要翻译或配音 -> 交给
video-minutes或直接 whisper。 - 只需把文字合成为音频、不涉及视频 -> 交给
mlx-tts。 - 需要保留原背景音乐/音效并做专业混音(人声分离)-> 本技能默认不做 Demucs,需另行处理。
- 视频无音轨或音质极差到无法识别。
Workflow
端到端推荐直接用编排脚本 <skill>/scripts/dub.py(见 ## Scripts),它会依次执行下面各步并
在翻译处暂停等待 AI 翻译。手动执行时按以下步骤:
-
抽音轨(16kHz 单声道):
ffmpeg -y -i input.mp4 -ar 16000 -ac 1 -c:a pcm_s16le audio16k.wav -
ASR 转录(要逐段准确时间戳):
whisper audio16k.wav --model large-v3-turbo --language <源语言> --output_format srt --output_dir .产出audio16k.srt,时间戳与原视频天然对齐。想要更好的中文识别可改装 Qwen3-ASR (pip install qwen-asr)并用<skill>/scripts/qwen3_asr.py。 -
翻译字幕(由 AI 完成):读 SRT,逐段翻译。每个 SRT 段输出一行译文,严格保持顺序。 先数清段数(如 27 段 -> 27 行),存为
translated.txt,继续前核对行数一致。 -
生成配音:
python3 <skill>/scripts/dub_segments.py audio16k.srt translated.txt dubbing.wav subtitle_synced.srt --lang <目标语言>脚本会:用首段作音色参考保证音色一致;保留原始时间戳让字幕与画面同步;只对溢出槽位 的段做变速(atempo 0.88–1.20);把相邻小于 1s 的间隔桥接起来让听感自然。 -
合并配音音轨到视频:
ffmpeg -y -i input.mp4 -i dubbing.wav \ -map 0:v:0 -map 1:a:0 \ -c:v copy -c:a aac -b:a 192k -shortest output_temp.mp4然后确认音轨已被替换:
ffprobe -v error -show_entries stream=codec_type -of csv=p=0 output_temp.mp4(应显示video, audio)。 -
烧录硬字幕:
python3 <skill>/scripts/burn_subtitles.py output_temp.mp4 subtitle_synced.srt output_final.mp4把译文字幕永久渲染进画面。 -
清理(可选):删除
audio16k.wav、audio16k.srt、translated.txt、dubbing.wav、subtitle_synced.srt、output_temp.mp4等中间产物。
Decision Rules
- 想一键跑完 -> 用
dub.py;只想重跑某一步(如已有 SRT)-> 用dub.py --srt existing.srt跳步。 - ASR 引擎:默认
--asr whisper(large-v3-turbo);中文识别要求高 ->--asr qwen3。 - 不需要硬字幕 ->
dub.py --no-burn(只到音轨合并)。 - 目标语言由
--target-lang决定(默认zh);dub_segments.py的 TTS 语言由--lang决定。 - 翻译必须由 AI 完成且与 SRT 段严格 1:1;行数不符时脚本直接报错退出,不要绕过。
- 字幕没出现/中文变方框 -> 用
burn_subtitles.py(moviepy 渲染,自动找中文字体),不要依赖 ffmpeg 的subtitles=(libass 缺失时会静默失败)。 - 配音时长溢出槽位 -> 依赖脚本的分段 atempo,不要对整条音轨做全局变速。
What ships with it
6 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.
- 1mo ago Changed · -7 lines · +161 tokens per session 031107fc74ac
- 2mo ago First seen · 143 lines · 66 tokens per session scan A e1b7ab942dd8
video-dubbing is a skill published in the GitHub repository renky1025/agent-skills (11 stars, last pushed 22d ago), licensed MIT. It adds 227 tokens to every session and 2,693 once invoked, about $0.0009 per session on Opus 5.5. 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 2026-08-31.
Other skills, from other repositories
iflytek-text-proofread
A tool that checks Chinese writing through iFlytek’s official document-proofreading service. It looks for spelling, punctuation, wording, factual, terminology, and sensitive-content problems.
iflytek-translate
A text-translation tool that uses iFlytek's machine-translation service and supports more than 70 language pairs. It can translate text supplied directly, through standard input, or from a file.
iflytek-video-translate
A tool that translates a video’s speech into another language and creates dubbed audio. Video localization means adapting video content for viewers who speak a different language.
edgeone skill scanner
A local static scanner that checks agent-skill files for security risks before they are installed or used. Static analysis examines files without running them.
babysit
Use when a user asks to babysit, monitor, keep checking, keep an eye on, or report when a pull request, CI run, ticket, deployment, or other changing target reaches an outcome.
rubber-duck
Adversarial "rubber duck" review that turns explaining-out-loud into a hallucination check. The main session is the PRESENTER (it did the work — a design doc, investigation, or analysis — and holds the real reasoning) and reconstructs the topic to a LISTENER — a spawned subagent pinned to a DIFFERENT-vendor model that…