← 返回 Skills 市场
287
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install chattts
功能描述
High-quality, conversational Text-to-Speech (TTS) generation via local ChatTTS API.
使用说明 (SKILL.md)
ChatTTS Voice Generator
AI-optimized conversational text-to-speech using a local ChatTTS server. Designed to generate highly natural human voices with realistic laughs, breathing, and pauses.
Generate Speech
node {baseDir}/scripts/tts.mjs "你要转换的语音文本"
node {baseDir}/scripts/tts.mjs "你好啊！[laugh] 今天天气真不错，[uv_break] 我们出去玩吧？" --seed 2048
node {baseDir}/scripts/tts.mjs "写代码真是太开心了！" --seed 1234 --temperature 0.5
Options
--seed \x3Cnumber>: Random seed to fix the speaker's voice/timbre (default: 2048). Change this number to switch between different male/female voices.--temperature \x3Cfloat>: Controls the emotional variance and stability (default: 0.3). Lower is more stable and clear, higher (e.g., 0.6) is more expressive but might mumble.--top_p \x3Cfloat>: Top P sampling parameter for voice generation (default: 0.7).
Notes
- Requires the local ChatTTS FastAPI server to be running (default target: http://172.23.252.114:8020).
- Ensure CHATTTS_API_URL is set in your .env file if the API is hosted on a different machine.
- PRO TIP: Always try to insert [laugh] (laughter) and [uv_break] (pauses/breaths) into the text to make the generated voice sound exactly like a real human.
- The script will return the absolute local file path of the generated .wav audio file.
安全使用建议
This skill appears to be a simple Node client for a ChatTTS server and mostly behaves as described, but there are configuration inconsistencies you should resolve before installing:
- Metadata issue: the registry lists a literal URL (http://172.23.252.114:8020) as a required env var and as the primary credential. That is likely a mistake; the script expects an environment variable named CHATTTS_API_URL. Ask the publisher to fix the metadata.
- Default endpoint: the script defaults to a private IP (172.23.252.114:8020). Verify that this host is one you control/trust. If not, set CHATTTS_API_URL to a trusted localhost or internal server before use.
- Port mismatch: the script's default port is 8020, but the error message mentions checking port 8080 — another sign of sloppy editing. Confirm which port your ChatTTS server actually uses.
- Privacy: any text you pass to the skill is POSTed to the configured API. Don’t send sensitive content unless you trust the API host.
Recommended actions: inspect or run the included scripts in an isolated environment; correct CHATTTS_API_URL in the skill metadata or set CHATTTS_API_URL explicitly to a known endpoint (e.g., http://127.0.0.1:8020) if you run a local server; request source/homepage or author verification if you need higher assurance. If the author fixes the metadata and confirms the endpoint is local/controlled, this would likely be coherent/benign.
功能分析
Type: OpenClaw Skill
Name: chattts
Version: 1.0.0
The skill provides Text-to-Speech functionality by sending text to a ChatTTS API. While it hardcodes a specific private IP address (172.23.252.114:8020) as the default endpoint in both SKILL.md and scripts/tts.mjs, this appears to be a local development configuration rather than a malicious SSRF or exfiltration attempt. The script logic is straightforward, lacks obfuscation, and aligns perfectly with the stated purpose of generating audio files from text.
能力评估
Purpose & Capability
The skill is a Node-based client that POSTs text to a ChatTTS FastAPI server and returns a local .wav path — this matches the stated TTS purpose and the requirement for the 'node' binary. However, the registry's 'required env vars' field contains a literal URL (http://172.23.252.114:8020) instead of an environment variable name (SKILL.md and the script expect CHATTTS_API_URL). That metadata mismatch is unexpected and likely a misconfiguration.
Instruction Scope
SKILL.md and scripts/tts.mjs limit actions to reading CHATTTS_API_URL (or using the built-in default), sending the provided text to /v1/audio/speech, and printing the generated file path. The script does not read arbitrary files or other env vars. Note: the script will send any provided text to the configured API endpoint — so the privacy/security of the input depends on where that endpoint points.
Install Mechanism
This is an instruction-only skill with a small included Node script and no install spec; nothing is downloaded or written by an installer. Risk from install mechanism is low.
Credentials
The declared required env var in registry metadata is a URL literal (http://172.23.252.114:8020) and primaryEnv is the same URL; this is malformed and disproportionate — expected: a single env var name (CHATTTS_API_URL). The script defaults to a private IP (172.23.252.114:8020), so by default user text will be sent to that host. While no secret keys are requested, sending arbitrary text to an unknown IP can expose sensitive content.
Persistence & Privilege
The skill is not marked always:true and doesn't request elevated persistence. It does not modify other skills or system-wide configs in the provided files.
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 chattts - 安装完成后，直接呼叫该 Skill 的名称或使用
/chattts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 of chattts skill: high-quality, conversational Text-to-Speech (TTS) generation via a local ChatTTS API.
- Supports configurable voice, expressiveness, and natural-sounding elements like laughs and breathing with text tags.
- Command-line script for TTS generation, allowing custom seed, temperature, and top_p parameters.
- Returns generated speech as a .wav audio file; requires local ChatTTS server.
元数据
常见问题
Chattts 是什么？
High-quality, conversational Text-to-Speech (TTS) generation via local ChatTTS API. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 287 次。
如何安装 Chattts？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install chattts」即可一键安装，无需额外配置。
Chattts 是免费的吗？
是的，Chattts 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Chattts 支持哪些平台？
Chattts 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Chattts？
由 Djttt（@djttt）开发并维护，当前版本 v1.0.0。
推荐 Skills