Byted Voice Notify

使用火山引擎API向指定手机号发送语音通知，可根据语言和关键词选择语音资源并自动调用号码池发送。

Security Scan

The skill's implementation matches its stated purpose (sending voice calls via Volcano/Volcengine) and only needs Volcengine API keys, but packaging metadata is inconsistent about required credentials which suggests sloppy or incomplete configuration that you should clarify before installing.

Purpose & Capability

The name/description, SKILL.md and the included script consistently implement sending voice notifications via Volcengine (cloud-vms.volcengineapi.com). However, the registry-level 'Requirements' block at the top of the provided metadata says 'Required env vars: none' while the SKILL.md metadata and the script both require VOLCENGINE_ACCESS_KEY and VOLCENGINE_SECRET_KEY — this metadata mismatch is inconsistent and should be resolved.

Instruction Scope

The SKILL.md gives narrow, concrete runtime instructions: extract a phone number, run the included Python script with 'pool', 'resource', or 'send', choose items from returned JSON, and report results. The instructions only reference the VOLCENGINE env vars and do not ask the agent to read other system files or call unexpected external endpoints.

Install Mechanism

There is no install spec (instruction-only style) and the skill includes a Python script bundled with the skill. No external downloads or package installs are performed, and the script uses standard library modules and direct HTTPS calls — low install risk.

Credentials

The Python script and SKILL.md legitimately require VOLCENGINE_ACCESS_KEY and VOLCENGINE_SECRET_KEY to authenticate to Volcengine, which is proportionate to the stated purpose. The problem is that the top-level registry metadata omitted these required env vars — a packaging inconsistency that could cause surprise at runtime and should be corrected. No other unrelated credentials are requested.

Persistence & Privilege

The skill does not request permanent 'always' inclusion, does not modify other skills or system-wide settings, and runs only the bundled script on demand. Autonomous model invocation is allowed by default but is not combined with other concerning privileges.

What to consider before installing

This skill appears to do what it says (send voice notifications via Volcengine) and its code is readable and uses the Volcengine API endpoint. However, the published metadata omitted the required environment variables while SKILL.md and the script do require VOLCENGINE_ACCESS_KEY and VOLCENGINE_SECRET_KEY — that's an inconsistency you should clear up before installing. Before you proceed: 1) Confirm with the publisher that the two VOLCENGINE env vars are required and that no other secrets are needed. 2) Only supply API keys with the minimum necessary permissions for sending voice notifications, and consider using scoped/temporary credentials. 3) Review the included script (it is bundled) and verify the host cloud-vms.volcengineapi.com is correct for your account. 4) Run it in an isolated environment or sandbox first (so accidental sends don't happen) and test with numbers you control or consent has been obtained. 5) If you need higher assurance, ask the publisher to fix the registry metadata so required envs are declared correctly.

火山引擎语音通知发送技能

基于火山引擎语音服务API向指定手机号码发送语音通知。

适用场景

1. 用户提到「给{手机号}发送语音通知」「帮我给{手机号}打语音电话」
2. 用户提到「用{语言}给{手机号}发送语音通知」
3. 用户提到「给{手机号}发送{关键词}语音通知」
4. 用户需要通过语音方式通知重要信息

核心执行流程

当你收到发送语音通知的请求时，请严格按照以下 4 个步骤依次执行：

---

第 1 步：提取手机号码

从用户的指令中提取 11 位手机号码。如果用户没有提供手机号，请先询问用户。

---

第 2 步：查询可用号码池

执行以下命令查询号码池：

python3 scripts/voice_notify.py pool

从返回的 JSON 结果中：

• 查看 Result.Records 数组
• 选择一个 NumberCount > 0（有号码可用）的号码池
• 提取该号码池的 NumberPoolNo 字段值（例如：NP165357678402826428）

注意： 优先选择第一个有号码的号码池。

---

第 3 步：查询语音资源

执行以下命令查询语音资源：

python3 scripts/voice_notify.py resource

从返回的 JSON 结果中：

• 查看 Result.Records 数组
• 根据用户指令中的关键词，在资源的 Name、Remark、TtsTemplateContent、Lang 字 段中匹配
• 选择最匹配的语音资源
• 提取该资源的 ResourceKey 字段值

资源选择规则：

1. 如果没有匹配的关键词，选择列表中的第一个资源

---

第 4 步：发送语音通知

执行以下命令发送语音通知：

python3 scripts/voice_notify.py send <手机号> <资源Key> <号码池编号>

将第 1 步提取的手机号、第 3 步提取的资源Key、第 2 步提取的号码池编号填入命令中。

---

第 5 步：返回结果

将 send 命令的返回结果直接告知用户。

---

脚本命令说明

命令	参数	说明
resource	无	查询可用语音资源列表
pool	无	查询语音通知号码池列表
send	<手机号> <资源Key> <号码池编号>	发送语音通知

完整使用示例

示例 1：简单发送

用户输入： 给13800138000发语音通知

你的执行：

1. 提取手机号：[phone]

2. 查询号码池：

python3 scripts/voice_notify.py pool

假设返回结果中第一个有号码的号码池的 NumberPoolNo 是 NPabc

3. 查询语音资源：

python3 scripts/voice_notify.py resource

假设返回结果中第一个资源的 ResourceKey 是 312346368e676406285b8463931f090f4

4. 发送语音通知：

python3 scripts/voice_notify.py send 13800138000 312346368e676406285b8463931f090f4 NPabc

5. 将发送结果告知用户

---

示例 2：指定语言

用户输入： 用日语给13800138000发送语音通知

你的执行：

1. 提取手机号：[phone]

2. 查询号码池：

python3 scripts/voice_notify.py pool

选择第一个有号码的号码池

3. 查询语音资源：

python3 scripts/voice_notify.py resource

在返回的资源列表中，查找 Lang 为 jap 或 Name/Remark 包含"日语"的资源，提取其 ResourceKey

4. 发送语音通知：

python3 scripts/voice_notify.py send 13800138000 <日语资源Key> <号码池编号>

5. 将发送结果告知用户

---

示例 3：指定关键词

用户输入： 给13800138000发送OCIC语音通知 你的执行：

1. 提取手机号：[phone]

2. 查询号码池：

python3 scripts/voice_notify.py pool

选择第一个有号码的号码池

3. 查询语音资源：

python3 scripts/voice_notify.py resource

在返回的资源列表中，查找 Name 或 Remark 包含"OCIC"的资源，提取其 ResourceKey

4. 发送语音通知：

python3 scripts/voice_notify.py send 13800138000 <OCIC资源Key> <号码池编号>

5. 将发送结果告知用户

---

环境变量

使用前需配置火山引擎访问密钥到系统环境变量或 OpenClaw 的 .env 文件：

• VOLCENGINE_ACCESS_KEY：火山引擎访问密钥ID
• VOLCENGINE_SECRET_KEY：火山引擎秘密访问密钥

获取方式：登录火山引擎控制台，在访问密钥管理中创建和获取。

错误处理

• 若报错 请设置环境变量 VOLCENGINE_ACCESS_KEY 和 VOLCENGINE_SECRET_KEY：提示用户配置火山引擎访问密钥。
• 若报错 没有找到可用的号码池：提示用户检查账户下是否有可用的语音通知号码池。
• 若报错 没有找到可用的语音资源：提示用户检查账户下是否有可用的语音资源文件。
• 遇到其他报错时直接告知用户具体错误信息。

参考文档

• https://www.volcengine.com/docs/6358/166389?lang=zh — 签名机制
• https://www.volcengine.com/docs/6358/1722078?lang=zh — 查询可用语音资源
• https://www.volcengine.com/docs/6358/172952?lang=zh — 单次发送语音通知
• https://www.volcengine.com/docs/6358/173339?lang=zh — 查询号码池列表