$ ~/registry/skill/openclaw-openai-whisper-api

SKILL

openai-whisper-api

Name: openai-whisper-api
Author: openclaw

Transcribe audio into text or speaker-separated transcripts with OpenAI APIs.

Stars

★ 382,582

Source

GitHub

Updated

2026-07-11

// securityCaution

Prompt only, no code
Open source
Community-trusted· 377.5k

Summary

Overall this is a low-complexity local skill that mainly calls a transcription endpoint via script; there are no clear signs of credential abuse, suspicious injection, or excessive system permissions. It does read local audio and may send content to the configured OpenAI-compatible endpoint, so it is cautionary rather than high risk.

CredentialsCaution

It requires OPENAI_API_KEY or reads apiKey from the OpenClaw config file; this is normal key usage, with no sign of extra credential harvesting or leakage in the material.

Network egressCaution

It uploads audio to /v1/audio/transcriptions and can use OPENAI_BASE_URL to target an OpenAI-compatible proxy or local gateway; user audio is sent out, but only to declared known endpoints.

Code executionCaution

The README shows local execution of scripts/transcribe.sh to perform transcription; this is a normal local process execution capability for a skill/tool, with no extra system permissions indicated.

Data accessCaution

The script reads the user-specified local audio file and defaults to writing .txt/.json outputs locally; access appears limited to the provided input and output files.

Supply chainCaution

The source is a GitHub open-source repository with very high star count, which reduces supply-chain concern; however, the license is undeclared and maintenance status is unknown, so treat it as auditable but worth caution.

Recommendations

Prefer a trusted OpenAI-compatible endpoint or a local gateway, and upload only the necessary audio.
Use least-privilege, dedicated API credentials instead of reusing high-privilege keys.
Inspect the script and output paths before running to ensure it won’t overwrite unexpected local files.
For higher assurance, review the repository license, recent commits, and dependency list.

Audited by: gpt-5.4 · 2026-06-17

// install

Copy the install command and let the AI configure it · recommended for beginners

Please install the "openai-whisper-api" skill from askskill:
1. Download https://raw.githubusercontent.com/openclaw/openclaw/main/skills/openai-whisper-api/SKILL.md
2. Save it as ~/.claude/skills/openai-whisper-api/SKILL.md
3. Reload skills and tell me it's ready

// download

Download SKILL.mdInstall manifest ↗

// examples

Convert meeting audio to text

Prompt

Use openai-whisper-api to transcribe this meeting recording into Chinese text, organize it into paragraphs, and preserve chronological order.

Expected

A readable Chinese meeting transcript organized in chronological order.

Diarize an interview recording

Prompt

Use openai-whisper-api to process this interview audio and return a diarized transcript with each segment labeled by speaker.

Expected

A segmented interview transcript with speaker labels.

Create transcript material from a podcast

Prompt

Use openai-whisper-api to transcribe this English podcast into English text, preserving proper nouns as much as possible for later summarization and content organization.

Expected

An accurate English podcast transcript ready for summarization or editing.

// docs

OpenAI transcriptions API

Transcribe audio through /v1/audio/transcriptions. Set OPENAI_BASE_URL for an OpenAI-compatible proxy or local gateway.

Quick start

{baseDir}/scripts/transcribe.sh /path/to/audio.m4a

Defaults:

Model: gpt-4o-transcribe
Output: <input>.txt

Useful flags

{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --model gpt-4o-transcribe --out /tmp/transcript.txt
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --model gpt-4o-mini-transcribe
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --model gpt-4o-transcribe-diarize --json
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --model whisper-1
{baseDir}/scripts/transcribe.sh /path/to/audio.m4a --language en
{baseDir}/scripts/transcribe.sh /path/to/audio.m4a --prompt "Speaker names: Peter, Daniel"
{baseDir}/scripts/transcribe.sh /path/to/audio.m4a --json --out /tmp/transcript.json

Notes:

Supported upload formats include mp3, mp4, mpeg, mpga, m4a, wav, webm.
25 MB upload limit on the hosted API.
Use diarize for speaker labels; script sends chunking_strategy=auto and rejects --prompt.

API key

Set OPENAI_API_KEY, or configure it in the active OpenClaw config file ($OPENCLAW_CONFIG_PATH, default ~/.openclaw/openclaw.json). Optionally set OPENAI_BASE_URL:

{
  skills: {
    "openai-whisper-api": {
      apiKey: "OPENAI_KEY_HERE",
    },
  },
}

openclawGet→

// similar tools

SKILL

★383k

openai-whisper

Transcribe audio to text locally with Whisper CLI, no API key required.

// examples

Convert meeting audio to text

Prompt

Use openai-whisper-api to transcribe this meeting recording into Chinese text, organize it into paragraphs, and preserve chronological order.

Expected

A readable Chinese meeting transcript organized in chronological order.

Diarize an interview recording

Prompt

Use openai-whisper-api to process this interview audio and return a diarized transcript with each segment labeled by speaker.

Expected

A segmented interview transcript with speaker labels.

Create transcript material from a podcast

Prompt

Use openai-whisper-api to transcribe this English podcast into English text, preserving proper nouns as much as possible for later summarization and content organization.

Expected

An accurate English podcast transcript ready for summarization or editing.

// docs

OpenAI transcriptions API

Transcribe audio through /v1/audio/transcriptions. Set OPENAI_BASE_URL for an OpenAI-compatible proxy or local gateway.

Quick start

{baseDir}/scripts/transcribe.sh /path/to/audio.m4a

Defaults:

Model: gpt-4o-transcribe
Output: <input>.txt

Useful flags

{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --model gpt-4o-transcribe --out /tmp/transcript.txt
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --model gpt-4o-mini-transcribe
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --model gpt-4o-transcribe-diarize --json
{baseDir}/scripts/transcribe.sh /path/to/audio.ogg --model whisper-1
{baseDir}/scripts/transcribe.sh /path/to/audio.m4a --language en
{baseDir}/scripts/transcribe.sh /path/to/audio.m4a --prompt "Speaker names: Peter, Daniel"
{baseDir}/scripts/transcribe.sh /path/to/audio.m4a --json --out /tmp/transcript.json

Notes:

Supported upload formats include mp3, mp4, mpeg, mpga, m4a, wav, webm.
25 MB upload limit on the hosted API.
Use diarize for speaker labels; script sends chunking_strategy=auto and rejects --prompt.

API key

Set OPENAI_API_KEY, or configure it in the active OpenClaw config file ($OPENCLAW_CONFIG_PATH, default ~/.openclaw/openclaw.json). Optionally set OPENAI_BASE_URL:

{
  skills: {
    "openai-whisper-api": {
      apiKey: "OPENAI_KEY_HERE",
    },
  },
}