← 返回 Skills 市场

Volcengine TTS to TOS Agent

Name: Volcengine TTS to TOS Agent
Author: day253

作者 day253 · GitHub ↗ · v1.0.0

cross-platform ⚠ suspicious

322

总下载

当前安装

版本数

在 OpenClaw 中安装

/install day253-volcengine-agent-tts-to-tos

功能描述

Combined agent that synthesizes speech via Volcengine TTS, uploads the audio to TOS, and returns a presigned temporary URL. Use when users need a shareable a...

使用说明 (SKILL.md)

Category: agent

TTS → TOS → Presigned URL Agent

一步完成：文本 → 语音合成 → 上传对象存储 → 生成临时访问链接。

Workflow

┌──────────┐     ┌──────────────┐     ┌──────────────┐     ┌──────────────┐
│  输入文本  │ ──▶ │  TTS 语音合成  │ ──▶ │  上传到 TOS   │ ──▶ │ 生成预签名 URL │
└──────────┘     └──────────────┘     └──────────────┘     └──────────────┘
                   audio bytes           tos://bucket/key       https://...

调用火山引擎 TTS HTTP API 合成音频（mp3/wav）
将音频上传到指定 TOS 桶
生成带有效期的预签名 URL
输出完整结果（本地路径 + TOS 路径 + 临时链接）

Prerequisites

pip install requests tos
TTS 环境变量：VOLCENGINE_TTS_APP_ID, VOLCENGINE_TTS_TOKEN, VOLCENGINE_TTS_CLUSTER
TOS 环境变量：VOLCENGINE_ACCESS_KEY, VOLCENGINE_SECRET_KEY, VOLCENGINE_TOS_ENDPOINT, VOLCENGINE_TOS_REGION

Quick start

# 基本用法：文本 + 目标桶
python agents/tts-to-tos/scripts/tts_to_tos.py \
  --text "你好，这是一段测试语音" \
  --bucket my-bucket

# 完整参数
python agents/tts-to-tos/scripts/tts_to_tos.py \
  --text "欢迎使用火山引擎语音合成服务" \
  --voice-type BV700_streaming \
  --encoding mp3 \
  --bucket my-bucket \
  --key-prefix audio/tts/ \
  --expires 7200 \
  --print-json

Parameters

参数	必填	默认值	说明
`--text`	✓		要合成的文本
`--bucket`	✓		TOS 目标桶名
`--voice-type`		`BV700_streaming`	音色
`--encoding`		`mp3`	音频格式 mp3/wav/pcm
`--speed-ratio`		`1.0`	语速 [0.2, 3]
`--volume-ratio`		`1.0`	音量 [0.1, 3]
`--pitch-ratio`		`1.0`	音高 [0.1, 3]
`--language`			语言，如 `cn`
`--key-prefix`		`tts/`	TOS 对象键前缀
`--expires`		`3600`	预签名 URL 有效期（秒）
`--keep-local`		`false`	保留本地临时音频文件
`--print-json`		`false`	输出 JSON 格式结果

Output

{
  "text": "你好，这是一段测试语音",
  "voice_type": "BV700_streaming",
  "duration_ms": "2150",
  "tos_bucket": "my-bucket",
  "tos_key": "tts/20260308-143025-a1b2c3d4.mp3",
  "presigned_url": "https://my-bucket.tos-cn-beijing.volces.com/tts/...",
  "expires_seconds": 3600,
  "audio_size": 34560
}

Safety

不执行任何删除操作
本地临时文件默认自动清理（除非 --keep-local）
预签名 URL 有效期可控，默认 1 小时

References

TTS skill: skills/ai/audio/volcengine-ai-audio-tts/SKILL.md
TOS skill: skills/storage/tos/volcengine-storage-tos/SKILL.md

安全使用建议

This package largely does what it says (synthesize TTS, upload to TOS, return a presigned URL), but there are two practical concerns you should address before installing: (1) metadata inconsistency — the registry says no env vars required but the script needs multiple Volcengine keys/tokens; verify you are comfortable providing those credentials and that the registry/package owner is trusted; (2) the script automatically loads .env files from your working directory and repository root, which can cause it to pick up unrelated secrets. Run it in an isolated environment (ephemeral VM or container), verify or sanitize any .env files, and supply least-privilege credentials (dedicated TOS bucket policy and short presigned expiry). Also review/confirm the source of the 'tos' Python package (where it is installed from) and test with non-production credentials first. If you need higher assurance, ask the publisher to correct the registry metadata to list required env vars and to expose an option to disable automatic .env loading.

功能分析

Type: OpenClaw Skill Name: day253-volcengine-agent-tts-to-tos Version: 1.0.0 The skill provides a legitimate utility for synthesizing speech using Volcengine TTS and uploading the resulting audio to Volcengine Object Storage (TOS). The Python script (scripts/tts_to_tos.py) follows the documented workflow, using official SDKs and API endpoints (openspeech.bytedance.com) with standard environment variable authentication, and contains no evidence of malicious intent, data exfiltration, or unauthorized execution.

能力评估

⚠ Purpose & Capability

The code and SKILL.md implement Volcengine TTS → upload to Volcengine Object Storage (TOS) → presigned URL, which matches the name/description. However the registry metadata claims no required environment variables or primary credential, while both SKILL.md and the script require multiple Volcengine TTS and TOS credentials (VOLCENGINE_TTS_APP_ID, VOLCENGINE_TTS_TOKEN, VOLCENGINE_TTS_CLUSTER, VOLCENGINE_ACCESS_KEY, VOLCENGINE_SECRET_KEY, VOLCENGINE_TOS_ENDPOINT, VOLCENGINE_TOS_REGION). This metadata mismatch is an incoherence that users should not ignore.

⚠ Instruction Scope

The SKILL.md instructions stay within the stated pipeline, but the included script loads .env files from the current working directory and from the repository root (it walks upward to a .git root). That behavior can cause the tool to read and use unrelated secrets stored in .env files without explicit prompting. The script also issues network calls (TTS HTTP API and TOS SDK) as expected for its purpose; no unexplained external endpoints were found beyond the Volcengine TTS endpoint and the configured TOS endpoint.

✓ Install Mechanism

There is no install spec (instruction-only skill with an included script). Dependencies are standard Python packages (requests and a tos SDK) and SKILL.md documents pip install requests tos. No arbitrary archive downloads or non-standard install locations were present in the files reviewed.

⚠ Credentials

The environment variables required by the script are proportional to the stated functionality (TTS and TOS credentials). However, the registry metadata incorrectly lists no required env vars. Additionally, the script's .env loading behavior can import any key/value from .env files found on disk, which increases the chance that unrelated secrets (e.g., other service keys) will be used by the process. The number and naming of variables requested (ACCESS_KEY, SECRET_KEY, TOKEN, etc.) are expected for this integration but should be explicitly declared by the package registry.

✓ Persistence & Privilege

The skill does not request permanent system-wide presence (always:false). It does not modify other skills or system-wide settings. It can be invoked autonomously by the agent (default), which is normal for skills; there is no evidence of privileged persistence or modification of other agents' configurations.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install day253-volcengine-agent-tts-to-tos
安装完成后，直接呼叫该 Skill 的名称或使用 /day253-volcengine-agent-tts-to-tos 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release: TTS synthesis → TOS upload → presigned URL pipeline

元数据

Slug day253-volcengine-agent-tts-to-tos

版本 1.0.0

许可证 —

累计安装 0

当前安装数 0

历史版本数 1

常见问题