About openai-whisper
A skill for local speech-to-text using the Whisper CLI without an API key. It is intended for executable transcription requests, not planning or conceptual discussions.
Key features
- 01Local speech-to-text
- 02Whisper CLI use
- 03No API key required
- 04Task-specific results and action notes
Use cases
- 01Transcribing speech locally
- 02Handling requests to use the Whisper CLI
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ cowork-os/ cowork-os --skill openai-whisperAfter installing, type/openai-whisperin Claude Code to confirm it loaded.
Details
From the project READMEOpenai-whisper
Purpose
Local speech-to-text with the Whisper CLI (no API key).
Routing
- Use when: Use when the user asks to local speech-to-text with the Whisper CLI no API key.
- Do not use when: Do not use when the request is asking for planning documents, high-level strategy, or non-executable discussion; use the relevant planning or design workflow instead.
- Outputs: Outcome from Openai-whisper: task-specific result plus concrete action notes.
- Success criteria: Returns concrete actions and decisions matching the requested task, with no fabricated tool-side behavior.
Trigger Examples
Positive
- Use the openai-whisper skill for this request.
- Help me with openai-whisper.
- Use when the user asks to local speech-to-text with the Whisper CLI no API key.
- Openai-whisper: provide an actionable result.
Negative
- Do not use when the request is asking for planning documents, high-level strategy, or non-executable discussion; use the relevant planning or design workflow instead.
- Do not use openai-whisper for unrelated requests.
- This request is outside openai-whisper scope.
- This is conceptual discussion only; no tool workflow is needed.
Runtime Prompt
- Current runtime prompt length: 380 characters.
- Runtime prompt is defined directly in
../openai-whisper.json.
Comments
Markdown supported