transcribe

transcribe is a skill for Claude Code, Codex from Desko77/cursor-1c-skills. It costs 196 tokens per session (7,187 once invoked), scanned A, a copy of transcribe, MIT.

A tool for turning video and audio recordings into written transcripts, with optional speaker separation and screen analysis. Speaker separation identifies which parts of the recording belong to different people.

In plain words
What is it for?
Use it to transcribe audio or video, identify speakers, summarize meetings, or inspect what appears on screen.
Why use it?
It removes the need to listen through recordings manually to recover their spoken content or meeting notes.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths; positional $N argument.

Good fit Use it to transcribe audio or video, identify speakers, summarize meetings, or inspect what appears on screen.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/desko77/cursor-1c-skills/transcribe
Install

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.

Any agent
npx skills add Desko77/cursor-1c-skills --skill transcribe
Clone the repo
git clone --depth 1 https://github.com/Desko77/cursor-1c-skills

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for transcribe

README.md
[![agentmods](https://agentmods.dev/badge/skills/desko77/cursor-1c-skills/transcribe/github.svg)](https://agentmods.dev/skills/desko77/cursor-1c-skills/transcribe)
Your own site
<a href="https://agentmods.dev/skills/desko77/cursor-1c-skills/transcribe"><img src="https://agentmods.dev/badge/skills/desko77/cursor-1c-skills/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.

agentmods 80×15 button for transcribe

Your own site · 80×15
<a href="https://agentmods.dev/skills/desko77/cursor-1c-skills/transcribe"><img src="https://agentmods.dev/badge/skills/desko77/cursor-1c-skills/transcribe.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 196 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,187 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
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 analysis-evasion · line 1
    Suspicious Unicode normalization or mixed-script content
    Fix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
How audits are shown
Origin 95% copy Near-identical to another mod in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00196 $0.07187
Opus 5 $0.00098 $0.03594
Sonnet 5 $0.00039 $0.01437
Haiku 4.5 $0.00020 $0.00719

Measured 6d ago against content hash f59f14d3a46c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

transcribe 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 6d ago.

The scan reads SKILL.md. This mod also ships 13 executable files (scripts/analyze_video_local.py, scripts/diarize_moss.py, scripts/diarize_sherpa.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Origin

This is a copy

95% identical to transcribe — 15 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

skills/transcribe/SKILL.md · 264 lines

How it starts

The opening of the file, as written. The whole thing — 264 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/transcribe - Транскрибация видео и аудио

Два движка:

  • Локальный (default для аудио): faster-whisper (CUDA) + опц. диаризация. Движок диаризации выбирается сам: без --num-speakers - pyannote community-1 (GPU, корректный автодетект числа спикеров, RTF ~0.064); с явным --num-speakers N - sherpa-onnx GPU (pyannote-segmentation-3.0 + eres2net, RTF ~0.24, точное N). Опция --diarize-engine moss - MOSS-Transcribe-Diarize end-to-end: ASR+диаризация одной моделью (без whisper-шага), лучше текст на технических терминах, но ~2x медленнее (RTF ~0.34), требует venv-moss (env MOSS_PYTHON). Нет затрат, не уходит наружу. ВИДЕО тоже можно разобрать полностью локально - --engine local (разбор экрана локальной VLM + спикеры по голосу, см. ниже).
  • Gemini (default для видео и --analyze-ui): облачный API, ~$0.10/час. Нужен интернет и квота. Стартовая модель gemini-2.5-flash (пин конкретной версии, дешевая); при перегрузке (503/429) переходит на gemini-2.5-flash-lite. Дорогие 3.5/pro сознательно исключены.

Выбор движка по умолчанию

Тип файла Движок Причина
Аудио (m4a, mp3, wav, ogg, flac, aac, wma) local Быстро, бесплатно, диаризация
Видео (mp4, mkv, webm, avi, mov) gemini Быстро, облако. Приватный вариант - --engine local (см. ниже)
Видео + --engine local local Разбор экрана БЕЗ облака: whisper + локальная VLM (LM Studio) + спикеры по голосу
Любой + --analyze-ui gemini Детальный разбор интерфейсов в облаке
Любой + --engine gemini gemini Явный override на облако
Аудио + --engine local local Явный override (аудио)

При 503/429 Gemini-движок сначала сам перебирает пул моделей (см. раздел "Авто-fallback по моделям Gemini"). Если весь пул недоступен и это аудио - можно вручную переключиться на local (--engine local).

Режимы

Локальный (аудио + faster-whisper + опц. pyannote)

Выходные файлы:

  • <имя> - транскрипция.md - таймкоды + текст
  • <имя> - транскрипция.txt - plain text
  • <имя> - со спикерами.md - реплики с метками [SPEAKER_XX, MM:SS] (только при --diarize)

Read the full file on GitHub · 264 lines

Changes

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.

  1. 6d ago First seen · 264 lines · 196 tokens per session scan A f59f14d3a46c

Subscribe to this mod's changes

transcribe is a skill published in the GitHub repository Desko77/cursor-1c-skills (55 stars, last pushed 7d ago), licensed MIT. It adds 196 tokens to every session and 7,187 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. It is 95% identical to transcribe, differing in 15 lines, and is treated as a copy.