
from dotfiles947
Step-by-step instructions to transcribe audio files (podcasts, MP3s, interviews) locally using OpenAI Whisper CLI, with model recommendations and output formats
Guides a developer through transcribing audio files locally using the OpenAI Whisper CLI. Includes how to obtain audio (download from a URL or find podcast MP3s), recommended model choices (tiny→large tradeoffs), example whisper command lines, and tips for long files and subtitle outputs.
Use this skill when asked to "listen to" or transcribe podcasts, interviews, or other audio files, especially when a local Whisper CLI workflow is preferred. Good for ad-hoc transcription, subtitle generation (srt), or quick drafts where privacy/local processing is desired.
Useful for developer/workflow agents and local automation tools that can run shell commands and manage files.
A clean, simple skill for transcribing audio files using OpenAI Whisper CLI. No bundled scripts — it's purely instructional, guiding the agent to download audio and run whisper commands. The SKILL.md is well-written with clear steps, model recommendations, and practical tips. No security concerns beyond downloading arbitrary URLs with curl.
Pure instruction skill with no scripts. Clean frontmatter, clear documentation. Very low security risk — only uses curl to download to file and whisper for local transcription. No network exfiltration or destructive commands.