AI Video Transcriber is an open-source tool that turns online videos, podcasts, and local media files into transcripts, summaries, and archived content. It is for people who need to extract and organize spoken content from services such as YouTube, TikTok, Bilibili, Apple Podcasts, and SoundCloud.
Borrowing it
Nothing to install: this file belongs to wendy7756/AI-Video-Transcriber. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/wendy7756/AI-Video-Transcriber/main/.claude/skills/video-transcribe/SKILL.mdgit clone --depth 1 https://github.com/wendy7756/AI-Video-TranscriberWrote 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/wendy7756/ai-video-transcriber/video-transcribe)<a href="https://agentmods.dev/skills/wendy7756/ai-video-transcriber/video-transcribe"><img src="https://agentmods.dev/badge/skills/wendy7756/ai-video-transcriber/video-transcribe/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/wendy7756/ai-video-transcriber/video-transcribe"><img src="https://agentmods.dev/badge/skills/wendy7756/ai-video-transcriber/video-transcribe.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, 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 Privilege Escalation · line 20 Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
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 | $0.00108 | $0.01022 |
| Opus 5 | $0.00054 | $0.00511 |
| Sonnet 5 | $0.00022 | $0.00204 |
| Haiku 4.5 | $0.00011 | $0.00102 |
Grade B, and why
video-transcribe scanned grade B with 1 finding 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- **No ffmpeg** → `brew install ffmpeg` (macOS) / `sudo apt install ffmpeg` (Debian/Ubuntu). Do not proceed without it; audio extraction will fail. How it starts
The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Video Transcribe
Runs this repo's pipeline headlessly: platform subtitles when available (seconds), local Whisper as fallback, then optimize → translate → summarize.
Before running
The command needs the project's virtualenv and ffmpeg. Verify once per session:
ls venv/bin/python && command -v ffmpeg
- No venv →
./install.sh(orpython3 -m venv venv && venv/bin/pip install -r requirements.txt) - No ffmpeg →
brew install ffmpeg(macOS) /sudo apt install ffmpeg(Debian/Ubuntu). Do not proceed without it; audio extraction will fail.
Run it
venv/bin/python transcribe.py "<URL or file path>" --json
Always pass --json — it puts machine-readable output on stdout and keeps progress
chatter on stderr. Parse the JSON rather than scraping the log lines.
Useful flags:
| Flag | When to use |
|---|---|
-l, --summary-language <code> |
Summary language: en, zh, es, fr, de, it, pt, ru, ja, ko, ar. Default en |
--no-llm |
No API key available, or the user only wants the raw transcript. Skips optimize/translate/summarize |
--no-video |
User doesn't want the source video kept (faster, less disk) |
--whisper-model small |
Accuracy matters more than speed. tiny→large, default base |
-o <dir> |
Write the Markdown somewhere other than ./temp |
The LLM steps need an OpenAI-compatible key via OPENAI_API_KEY (and optionally
OPENAI_BASE_URL). Without one the pipeline still transcribes but falls back to
basic formatting — prefer --no-llm in that case and summarize the transcript
yourself, which is usually better anyway since you have the full context.
Provider settings:
export OPENAI_API_KEY="your_api_key_here"
export OPENAI_BASE_URL="https://openrouter.ai/api/v1"
export OPENAI_TRANSLATION_MODEL="gpt-4o" # optional
For a one-off run, transcribe.py also accepts --api-key, --base-url, and
--model. Prefer environment variables when possible so API keys are not written
to shell history.
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 · 90 lines · 108 tokens per session scan B 73366e353239
video-transcribe is a skill published in the GitHub repository wendy7756/AI-Video-Transcriber (3,272 stars, last pushed 19d ago), licensed Apache-2.0. It adds 108 tokens to every session and 1,022 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
bilibili-upload
A workflow for using the sau command-line tool to log in to Bilibili, a Chinese video-sharing platform, check an account, and upload videos.
vugola
Clip any YouTube or video URL into short-form viral clips using Vugola AI, then auto-schedule them to TikTok, Instagram, and YouTube Shorts. Use when someone says "clip this video", "clip and schedule", "make clips from this URL", or wants short-form content from a long video. Handles the full pipeline end-to-end …
douyin-upload
A workflow for using the sau command-line tool to log in to Douyin, the Chinese short-video platform, check account cookies, and publish videos or image-and-text posts.
kuaishou-upload
A workflow for using the sau command-line tool to log in to Kuaishou, a Chinese social media platform, check account cookies, and publish videos or image-and-text posts.
xiaohongshu-upload
A workflow for using the sau command-line tool to log in to Xiaohongshu, a Chinese social media platform, check account cookies, and publish videos or image-and-text posts.
listen
Nested swiss-knife reference for local audio analysis — transcribe speech with Whisper, or extract musical features (tempo, key, dynamics, spectral profile) with librosa. Both run on the user's machine with no API key. Read this when the human asks you to transcribe a voice note, extract lyrics from singing, critique…