Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Tts Voice Generator

v1.3.0

文本转语音生成工具,支持浏览声音列表、选择声音、上传自定义音频(带文本内容)并命名保存、生成语音和查询任务状态等功能。

0· 107·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for seanfeng1234/tts-voice-generator.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Tts Voice Generator" (seanfeng1234/tts-voice-generator) from ClawHub.
Skill page: https://clawhub.ai/seanfeng1234/tts-voice-generator
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install tts-voice-generator

ClawHub CLI

Package manager switcher

npx clawhub@latest install tts-voice-generator
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
high confidence
!
Purpose & Capability
The name/description (文本转语音) matches the code: it calls TTS endpoints, lists voices, uploads audio and submits jobs to https://www.datamass.cn/ai-back. However the skill actually requires a datamass API key stored in ~/.openclaw/config.json (and optionally a base URL) even though the manifest declares no required env vars or config paths. That mismatch between declared requirements and actual needed configuration is an incoherence the user should know about.
!
Instruction Scope
SKILL.md and the runtime code explicitly instruct/expect the user to place an API key in ~/.openclaw/config.json and to provide local audio file paths for upload. The code opens that config file and will read local files passed to upload functions. These behaviors are in-scope for a TTS skill, but the instructions grant the skill access to a local config file and to any local file path the user supplies — and the manifest did not advertise that. There's also some minor implementation inconsistencies (different default workflow_id values in different files) which may cause unexpected behavior.
Install Mechanism
No install spec is present (instruction-only / Python code included). The only external dependency is requests (requirements.txt). No remote binary downloads or archive extraction are used. This is low install risk.
!
Credentials
The code requires a datamass API key (datamass_api_key) in ~/.openclaw/config.json to call remote APIs, but the registry metadata lists no required credentials or config paths. That omission is a proportionality/information mismatch. Additionally, the skill will upload user-provided audio to a third-party service (datamass.cn) and save custom voices to that service's database — a privacy/secret-exposure risk if users upload sensitive voice samples or others' voices.
Persistence & Privilege
The skill does not request always:true and does not declare elevated platform privileges. It will communicate with external APIs and save uploaded audio on the remote service, but it does not modify other skills or system-wide agent settings. Autonomous invocation (disable-model-invocation:false) is the platform default and not flagged by itself.
What to consider before installing
Things to consider before installing: - The skill requires an API key for datamass.cn stored in ~/.openclaw/config.json, but the registry metadata does not declare this—expect to add that file before use. Verify you are comfortable storing the API key there. - The skill uploads local audio files you provide to https://www.datamass.cn and the service will persist custom voices. Do not upload sensitive or third-party voice recordings unless you trust that service and its privacy/terms. - There is no homepage or source provenance in the manifest; confirm that datamass.cn is the intended backend and that you trust it before giving an API key. - The code is largely coherent with its stated TTS purpose, but there are small implementation inconsistencies (different workflow_id values and slightly different return handling across duplicated files) — test in a sandbox first. - If you need stricter guarantees: inspect/run the code in an isolated environment, or replace the default BASE_URL with a trusted endpoint you control. If you accept the privacy implications and provide an API key, the skill appears to perform the TTS functionality it advertises.

Like a lobster shell, security has layers — review code before you run it.

latestvk975hzbys4art90mmydz01543x849twf
107downloads
0stars
4versions
Updated 3w ago
v1.3.0
MIT-0

TTS 语音生成器

完整工作流程

┌─────────────┐    ┌──────────────────────┐    ┌─────────────┐    ┌─────────────┐
│ 1. 浏览声音  │ -> │ 2. 选择或上传声音     │ -> │ 3. 生成语音  │ -> │ 4. 获取结果  │
└─────────────┘    └──────────────────────┘    └─────────────┘    └─────────────┘

步骤说明:

  1. 浏览声音:查看可用声音列表
  2. 选择声音
    • 如果有合适的声音 → 说"用第 X 个声音"或"用 [声音名称]"
    • 如果没有合适的 → 上传自己的音频(需提供音频文件、声音名称、音频文本内容)
  3. 生成语音:提供文本内容,使用选中的声音生成
  4. 获取结果:等待任务完成,返回音频 URL

示例命令

1. 浏览可用声音

列出所有可用的声音

输出示例:

📋 可用声音列表:

  1. 派蒙 - 平静 (定制声音)
     试听文本:既然罗莎莉亚说足迹上有元素力...
  2. 龙某洋 (定制声音)
     试听文本:三十、功名尘与土八千里路云和月。
  ...

💡 使用方式:
  • 选择声音:说'用第 X 个声音'或'用 [声音名称]'
  • 上传自定义声音:说'上传我的声音'并提供音频文件、名称和音频内容

共 10 个声音

2. 选择已有声音生成

用第 1 个声音生成:你好,这是测试文本

用派蒙生成:春眠不觉晓,处处闻啼鸟

3. 上传自定义声音

格式:

上传我的声音:[文件路径],名字叫"[声音名称]",音频内容是"[音频文本]"

示例 1 - 本地文件路径:

上传我的声音:C:/audio/my_voice.wav,名字叫"温柔女声",音频内容是"春眠不觉晓,处处闻啼鸟"

示例 2 - 相对路径:

上传这个音频文件作为参考声音:./sample.wav,命名为"我的声音",文本内容是"白日依山尽,黄河入海流"

说明:

参数说明要求
音频文件WAV 或 MP3 格式建议 10-60 秒,一句话或一句诗词的录音
声音名称给声音起的名字如"温柔女声"、"成熟男声"等
音频内容音频中朗读的文本用于音色训练参考,必须与音频内容一致

4. 使用刚上传的声音生成

用温柔女声生成:欢迎使用我们的服务

5. 查询任务状态

查询任务状态:taskId_xxx

触发方式

当用户请求以下内容时触发:

类型示例
生成语音"生成语音" / "TTS" / "文本转语音"
浏览声音"列出声音" / "获取声音列表" / "有哪些声音"
上传声音"上传音频" / "添加声音" / "自定义声音" / "上传我的声音"
查询状态"查询任务状态" / "检查 TTS 进度"
选择声音生成"用第 X 个声音生成:[文本]" / "用 [声音名称] 生成:[文本]"
直接转换提供文本内容要求转换成语音

配置要求

首次使用

需要在 ~/.openclaw/config.json 中配置 API Key(与下载工具 Skill 共用):

{
    "datamass_api_key": "您的 API Key"
}

获取 API Key

  1. 访问 https://www.datamass.cn
  2. 注册并登录
  3. 创建 API Key
  4. 复制生成的 API Key 到配置文件中

功能特点

  • ✅ 先浏览声音列表,再决定是否上传自定义声音
  • ✅ 支持按序号选择声音("用第 2 个声音")
  • ✅ 支持按名称选择声音("用温柔女声")
  • ✅ 上传自定义音频时一步保存:音频文件 + 声音名称 + 音频文本
  • ✅ 异步任务处理,无需等待
  • ✅ 自动状态轮询
  • ✅ 音频文件保存到 OSS,24 小时有效
  • ✅ 自定义声音永久保存到数据库,可重复使用

核心功能

方法说明
start_tts_workflow()启动引导式交互流程
show_voice_list()显示可用声音列表(格式化输出)
list_voices()获取可用声音列表(原始数据)
select_voice(index)根据序号选择声音
generate_speech()提交 TTS 任务(支持 voice_index/voice_name/media_path)
check_task_status()查询任务状态
upload_audio()上传音频文件(支持 style_name 和 audio_text 参数)
add_custom_voice()添加定制声音(保存到数据库)
upload_and_add()一键上传并添加为定制声音(上传时同时保存音频文本)

注意事项

计费说明

  • TTS 服务按生成音频时长计费
  • 最低余额门槛:5 积分
  • 100 字以内:2 积分,超出部分每 100 字加 1 积分
  • 需要充值积分后使用

上传要求

  • 音频格式:WAV 或 MP3
  • 音频时长:建议 10-60 秒
  • 音频内容:一句话或一句诗词的清晰录音
  • 声音名称:自定义,用于后续调用
  • 音频文本:必须与音频中朗读的内容一致,用于音色训练参考

使用提示

  • 自定义声音上传后永久保存到数据库,可重复使用
  • 建议上传前先浏览可用声音列表,避免重复上传
  • 生成语音时可指定声音序号或名称
  • 任务提交后自动轮询状态,无需手动查询

Comments

Loading comments...