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 agentmods add skills/xfwang522/vidscribe/agent-skillnpx skills add XFWang522/vidscribe --skill agent-skillgit clone --depth 1 https://github.com/XFWang522/vidscribeWrote 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/xfwang522/vidscribe/agent-skill)<a href="https://agentmods.dev/skills/xfwang522/vidscribe/agent-skill"><img src="https://agentmods.dev/badge/skills/xfwang522/vidscribe/agent-skill.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00056 | $0.00915 |
| Opus 5 | $0.00028 | $0.00458 |
| Sonnet 5 | $0.00011 | $0.00183 |
| Haiku 4.5 | $0.00006 | $0.00092 |
Grade A, and why
vidscribe 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 3d 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Video Transcription (vidscribe)
Transcribe any video URL to text using cloud ASR APIs or a local Whisper model.
Prerequisites
System tools (must be installed):
yt-dlp: downloads audio from 1000+ sites (pip install yt-dlp)ffmpeg: audio format conversion (brew install ffmpeg/apt install ffmpeg)
Python package:
vidscribe: the transcription tool (pip install vidscribe)
If vidscribe is not installed, install it first:
pip install vidscribe
Provider Setup
vidscribe supports 5 ASR providers. At least one must be configured via environment variables in ~/.zshrc or ~/.bashrc:
| Provider | Env Vars | Best For |
|---|---|---|
| Volcengine (火山引擎) | VOLC_APP_KEY, VOLC_ACCESS_KEY |
Chinese content, cheapest |
| OpenAI Whisper | OPENAI_API_KEY |
Multilingual, best quality |
| Aliyun (阿里云) | DASHSCOPE_API_KEY |
Chinese, free trial |
| Deepgram | DEEPGRAM_API_KEY |
Fastest, $200 free credit |
| Local (faster-whisper) | None (install: pip install vidscribe[local]) |
Offline, free |
If no provider credentials are found, ask the user which provider they want to use and help them set it up.
Usage
# Auto-detect provider from environment variables
vidscribe "<VIDEO_URL>" -o "<OUTPUT_PATH>"
# Specify provider explicitly
vidscribe "<VIDEO_URL>" -p <provider> -o "<OUTPUT_PATH>"
# Language hint for non-Chinese/English content
vidscribe "<VIDEO_URL>" -l ja-JP -o "<OUTPUT_PATH>"
Parameters:
<VIDEO_URL>: Any video URL (Bilibili, YouTube, Douyin, Twitter, TikTok, Vimeo, etc.)-o <OUTPUT_PATH>: Where to save the transcript (omit to print to stdout)-p <provider>: Force a specific provider (volcengine,openai,aliyun,deepgram,local)-l <LANG>: Language hint (e.g.ja-JP,ko-KR,de-DE)--keep-audio: Also save the downloaded audio file
Workflow
The tool automatically:
- Downloads audio via yt-dlp and converts to MP3
- Uploads to a temporary host (cloud providers only) for the ASR API to access
- Sends to the configured ASR provider for transcription
- Outputs the full transcript text
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.
- 3d ago First seen · 97 lines · 56 tokens per session scan A 5fa7f8fa82d9
vidscribe is a skill published in the GitHub repository XFWang522/vidscribe (4 stars, last pushed 5mo ago), licensed MIT. It adds 56 tokens to every session and 915 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
delive-transcript-analyzer
Analyze, summarize, and extract insights from DeLive transcription sessions. Use when: user mentions DeLive, transcription, meeting transcripts, live captions, audio transcription, AI correction, corrected transcript, or transcript analysis; user wants to search, retrieve, summarize, correct, or process recorded…
yw-transcribe
Transcribes a local video or audio file into a traceable Chinese or mixed Chinese-English package with TXT, optional timestamps and SRT/VTT, glossary handling, and quality evidence. Use for 视频转文字, 音频转写, 逐字稿, 字幕, word timestamps, 本地模型选择, 转写环境配置, 专业术语保留, StepFun/阶跃/StepAudio cloud ASR, or Volcengine/豆包/Seed ASR. Not for…
whisper
OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual…
whisper
OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual…
whisper
OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual…
whisper
OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual…