技能详情（站内镜像，无评论）
许可证：MIT-0
MIT-0 ·免费使用、修改和重新分发。无需归因。
版本：v1.0.0
统计：⭐ 0 · 330 · 2 current installs · 2 all-time installs
⭐ 0
安装量（当前） 2
🛡 VirusTotal ：良性 · OpenClaw ：良性
Package：mh-openai-whisper
安全扫描（ClawHub）
- VirusTotal ：良性
- OpenClaw ：良性
OpenClaw 评估
The skill's claims (local Whisper CLI transcription, no API key) match what it requires and instructs; nothing requested appears disproportionate, though there are two small notes to verify before installing.
目的
Name/description align with requirements: the skill only needs the 'whisper' binary and provides brew install metadata for it. No unrelated credentials, binaries, or config paths are requested. Minor note: SKILL.md states the default model is 'turbo', which is unexpected for Whisper model naming and may be a documentation inaccuracy to confirm.
说明范围
Runtime instructions are narrowly scoped: example commands operate on a user-supplied audio file and mention models downloading to ~/.cache/whisper. This implies network access and disk writes to the user home directory (model cache) on first run — normal for local transcription but worth noting. Instructions do not ask for unrelated files, secrets, or external endpoints.
安装机制
Install uses a Homebrew formula (openai-whisper) which is a low-to-moderate risk install mechanism compared with arbitrary downloads. The SKILL metadata does not declare the tap or provenance of the formula; confirm the formula is from a trusted tap (Homebrew core or a reputable maintainer) before installing.
证书
No environment variables, credentials, or config paths are requested — proportional to a local CLI transcription tool.
持久
Skill does not request always-on or elevated privileges and is not modifying other skills or global agent settings. Its runtime effects are limited to installing/using the whisper binary and caching models under the user's home directory.
综合结论
This skill appears coherent and limited in scope, but before installing: 1) confirm the Homebrew formula 'openai-whisper' comes from a trusted tap (to avoid running untrusted code via brew); 2) be aware that Whisper will download model files to ~/.cache/whisper (network usage and disk space); 3) verify the SKILL.md model name ('turbo') if you need a specific model — it may be a documentation typo. If those checks look good, the skill's request…
安装（复制给龙虾 AI）
将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。
请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「MH openai-whisper」。简介：使用Whisper CLI实现本地语音转文本（无API密钥）。。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/mohdalhashemi98-hue/mh-openai-whisper/SKILL.md
（来源：yingzhi8.cn 技能库）SKILL.md
---
name: openai-whisper
description: Local speech-to-text with the Whisper CLI (no API key).
homepage: https://openai.com/research/whisper
metadata:
{
"openclaw":
{
"emoji": "🎙️",
"requires": { "bins": ["whisper"] },
"install":
[
{
"id": "brew",
"kind": "brew",
"formula": "openai-whisper",
"bins": ["whisper"],
"label": "Install OpenAI Whisper (brew)",
},
],
},
}
---
# Whisper (CLI)
Use `whisper` to transcribe audio locally.
Quick start
- `whisper /path/audio.mp3 --model medium --output_format txt --output_dir .`
- `whisper /path/audio.m4a --task translate --output_format srt`
Notes
- Models download to `~/.cache/whisper` on first run.
- `--model` defaults to `turbo` on this install.
- Use smaller models for speed, larger for accuracy.