// Transcribe local audio files with the OpenAI Audio Transcriptions API without adding OpenAI dependencies to the target project.
| name | openai-transcribe |
| description | Transcribe local audio files with the OpenAI Audio Transcriptions API without adding OpenAI dependencies to the target project. |
| allowed-tools | Bash(python *) |
| argument-hint | ["audio-file ..."] |
Use scripts/transcribe.py to transcribe local audio through the
OpenAI Audio Transcriptions API.
The script intentionally uses only the Python standard library. It
does not add openai or any other dependency to the repository being
worked on.
The script reads the API key in this order:
OPENAI_API_KEY from the current environment.--env-file /path/to/.env, reading OPENAI_API_KEY=....--auto-env, scanning local project .env files under
/home/alexey/git and using the first file that contains
OPENAI_API_KEY.The key is never printed.
python /home/alexey/git/.claude/skills/openai-transcribe/scripts/transcribe.py \
--auto-env \
--model whisper-1 \
--language ru \
--out-dir docs/source-recordings \
docs/source-recordings/example.m4a
For whisper-1, the script requests verbose_json and writes both
the raw JSON response and a timestamped .txt file.
Files larger than the OpenAI upload limit are split with ffmpeg
before transcription, then combined back into one timestamped output.
Fetch and transcribe Google Recorder voice notes. Use when the user shares a recorder.google.com link and wants the original audio file, a transcript, or wants to act on a voice note.
Initialize a new Python library with modern tooling, packaging, tests, and optional CLI support. Use when the user wants to scaffold a new Python package.
Add the standardized CI publish workflow (`.github/workflows/publish.yml` + `make release`) to an existing Python project so PyPI releases happen on tag push. Use when a project still publishes via a local script (`publish.py`, `hatch publish`, `twine upload`) or has no automated publish at all.
Release the current project to its package registry and GitHub by bumping the version, pushing a tag, and letting CI publish. Works for any project (Python/PyPI, Rust/crates.io, Node/npm, etc.) that has a CI publish workflow keyed off `v*` tags.
Run and fix stylint checks for prose, Markdown, lessons, workshops, docs, and agent-written text. Use when editing written content or when the user asks for style, polish, or lint cleanup.
Fetch YouTube video transcripts. Use when the user asks to get a YouTube video transcript, subtitles, or captions, or wants to analyze/summarize a YouTube video.