Voice Translator

说中文出外语语音——按住说中文，2-3秒内播放英/日/韩语音。支持场景模式、双向对话、常用句收藏。

This skill is coherent with its stated goal (it sends audio to senseaudio.cn using the SENSEAUDIO_API_KEY), but there are a few inconsistencies to consider before installing: (1) The SKILL.md expects you to plug in an LLM (e.g., Claude) for 'scene-aware' translation but it does not declare any env var for that — be prepared to provide additional API keys and understand where those requests will go. (2) The example playback commands use utilities like 'play' (or platform tools such as 'afplay') that are not listed as required; the skill may fail or try to call other programs. (3) Audio data is uploaded to api.senseaudio.cn — verify the privacy policy and trustworthiness of that endpoint and the provided homepage. (4) Because this is instruction-only, there is no bundled code reviewable beyond SKILL.md; if you need stronger assurance, ask the publisher for details about the LLM integration they expect and request explicit declarations of all required binaries and env vars, or run the skill in an isolated environment.

Type: OpenClaw Skill Name: voice-translator Version: 1.0.0 The skill is a legitimate voice translation tool that interfaces with the SenseAudio API (senseaudio.cn) for speech-to-text and text-to-speech services. The provided Python code in SKILL.md implements standard audio processing workflows, including temporary file handling and playback via system utilities like afplay or mpg123. No indicators of data exfiltration, malicious command execution, or prompt injection were found; the use of subprocess.run is restricted to hardcoded audio player commands.