agentleFS
Sign inStart free
lordmaculordmacu/nexo-rs/skills/openai-whisper/SKILL.md

OpenAI Whisper

Skill8 starsChanged 7 months agoApache-2.0View on GitHub

Transcribe audio files (≤25 MB) to text via OpenAI-compatible Whisper API.

  • Reads credentials

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 1 is distinctive, and 9 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 9

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this skill to transcribe audio files into text.

    line 12
  2. Distinctive:

    No arguments. Returns endpoint, default model, token presence, max file size.

    line 33
  3. Very specific:

    Default response_format: "text" for plain transcripts.

    line 47
  4. Very specific:

    Use verbose_json when the user needs timestamps or segment data (subtitles, alignment, search).

    line 48
  5. Very specific:

    Use srt / vtt when the user explicitly wants subtitle files.

    line 49
  6. Very specific:

    Always pass language if the audio language is known — accuracy improves significantly.

    line 50
  7. Very specific:

    Use prompt to feed proper nouns, jargon, or expected style ("Bible reading", "casual Spanish", "medical interview").

    line 51
  8. Very specific:

    If -32014 (payload too large) → ask user to compress audio (ffmpeg -i in.mp3 -ac 1 -ar 16000 out.mp3).

    line 52
  9. Very specific:

    If -32015 (unsupported media) → file format unsupported by provider.

    line 53
  10. Very specific:

    If -32011 (unauthorized) → WHISPER_OPENAI_API_KEY missing or invalid.

    line 54

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

OpenAI Whisper

Use this skill to transcribe audio files into text. Backed by the openai-whisper extension which calls any OpenAI-compatible /audio/transcriptions endpoint (OpenAI, Groq, local whisper.cpp).

Use when

  • "Transcribe this audio"
  • "Convert this voice note to text"
  • "Get subtitles from this recording"
  • User attaches .mp3, .wav, .m4a, .webm, .ogg and wants the text

Do not use when

  • Text-to-speech (TTS) — different skill / extension
  • Real-time streaming transcription — this is batch only
  • Speaker diarization with strict attribution — Whisper does not separate speakers reliably
  • Audio > 25 MB — split or compress (downsample to 16 kHz mono) before calling

Tools

status

No arguments. Returns endpoint, default model, token presence, max file size.

transcribe_file

  • file_path (string, required) — absolute or relative path, ≤ 25 MB
  • model (string, optional) — override default (whisper-1, whisper-large-v3, etc.)
  • language (string, optional) — ISO 639-1 hint (en, es, pt); improves accuracy
  • prompt (string, optional) — biases vocabulary/style; useful for technical terms or names
  • response_format (string, optional) — text (default) | json | verbose_json (with segments+timestamps) | srt | vtt
  • temperature (number, optional, 0..1) — 0 for deterministic

Returns {file_path, bytes, model, language, response_format, transcript: {text, ...}}.

Execution guidance

  • Default response_format: "text" for plain transcripts.
  • Use verbose_json when the user needs timestamps or segment data (subtitles, alignment, search).
  • Use srt / vtt when the user explicitly wants subtitle files.
  • Always pass language if the audio language is known — accuracy improves significantly.
  • Use prompt to feed proper nouns, jargon, or expected style ("Bible reading", "casual Spanish", "medical interview").
  • If -32014 (payload too large) → ask user to compress audio (ffmpeg -i in.mp3 -ac 1 -ar 16000 out.mp3).
  • If -32015 (unsupported media) → file format unsupported by provider; convert to mp3/wav.
  • If -32011 (unauthorized) → WHISPER_OPENAI_API_KEY missing or invalid.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free