← 返回 Skills 市场
775
总下载
2
收藏
0
当前安装
10
版本数
在 OpenClaw 中安装
/install zvukogram
功能描述
Text-to-Speech via Zvukogram API with SSML support. Use when you need to generate speech from text, create podcasts, voice notifications, or work with audio....
安全使用建议
This skill is coherent and appears to be a straightforward Zvukogram TTS client. Before installing/use: (1) Verify you trust https://zvukogram.com and are comfortable giving your Zvukogram token and account email to the tool; the scripts send them to the official API. (2) If you plan to merge audio, install ffmpeg from a trusted package source — merge.py invokes ffmpeg via subprocess but ffmpeg is not declared as a required binary. (3) Storing the token in ~/.config/zvukogram/config.json is convenient but ensure file permissions are appropriate (restrict to your user). (4) If you need higher assurance, inspect or run the included Python scripts in a sandbox to observe network calls (they use urllib to post to zvukogram.com) and confirm no unexpected endpoints are contacted.
功能分析
Type: OpenClaw Skill
Name: zvukogram
Version: 1.1.6
The zvukogram skill bundle is a legitimate integration for the Zvukogram Text-to-Speech API. It contains functional Python scripts for checking account balances (balance.py), generating speech (tts.py), and merging audio files via ffmpeg (merge.py). The bundle includes extensive documentation and SSML references (SSML.md, API.md) designed to help an AI agent use the service correctly, with no evidence of malicious intent, data exfiltration, or prompt injection attacks.
能力评估
Purpose & Capability
Name/description (Zvukogram TTS) matches the code and docs. Required env vars (ZVUKOGRAM_TOKEN, ZVUKOGRAM_EMAIL) are appropriate for the stated API usage. The scripts call only the documented zvukogram.com API endpoints.
Instruction Scope
SKILL.md and scripts focus on generating TTS, chunking, merging audio, and checking balance. The runtime instructions read a local config (~/.config/zvukogram/config.json) or environment variables — which is expected for an API client. They do not ask for unrelated files, credentials, or network endpoints.
Install Mechanism
There is no install spec (instruction-only), which is low-risk. The package includes Python scripts that use the stdlib (urllib) and a subprocess call to ffmpeg for merging. The skill does not declare ffmpeg as a required binary, yet examples and merge.py rely on it — a small documentation mismatch (not evidence of maliciousness) that users should note before running merging steps.
Credentials
Only two environment credentials are requested (token and email) which are proportionate for an API client. No other secret names, unrelated cloud creds, or system-wide tokens are required.
Persistence & Privilege
always: false and the skill does not request to persist or modify other skills or system-wide settings. It only suggests storing credentials in a local config file, which is normal for a client script.
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 zvukogram - 安装完成后，直接呼叫该 Skill 的名称或使用
/zvukogram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
v1.1.6
Sync latest production learnings into the public skill: explicitly document no-silent-truncation for >1000-char /text limits, strengthen SSML pipeline guidance to preserve useful inline tags when downstream runtimes support them, and keep wrapper-tag handling explicit. Includes prior enriched API/SSML/say-as/pronunciation/podcast references.
v1.1.5
Expand SSML and API references: add say-as coverage (fraction/date/time/telephone/currency/cardinal/ordinal/spell-out), chunking vs /text vs /longtext guidance, pronunciation patterns, podcast examples, and refreshed troubleshooting/docs links.
v1.1.4
SSML reference + cheatsheet update; docs refresh; env-friendly creds for isolated runs
v1.1.3
Credential handling for isolated runs (env vars); docs refresh
v1.1.2
Fix: Unify config path to ~/.config/zvukogram/config.json
v1.1.1
Fix: Add language info - optimized for Russian with stress marks, service supports multiple languages. Corrects version sequence after 1.0.2/1.1.0 mismatch.
v1.0.2
Fix registry metadata: add required env vars ZVUKOGRAM_TOKEN and ZVUKOGRAM_EMAIL, add credential zvukogram_api
v1.1.0
Zvukogram 1.1.0
- Reworked documentation for clarity and readability; SKILL.md is now in English with improved structure.
- Updated configuration instructions (now recommends ~/.config/zvukogram/config.json).
- Enhanced SSML usage examples and references.
- Improved examples and troubleshooting guides in the reference files.
- No functional changes to scripts, but documentation for usage and features is clearer.
v1.0.1
- Добавлена поддержка переменных окружения ZVUKOGRAM_TOKEN и ZVUKOGRAM_EMAIL для авторизации.
- В SKILL.md теперь явно указаны требования к API-токену и email, а также доступен альтернативный способ конфигурации через переменные окружения.
- В разделе meta-описания добавлено поле requires (env, credentials).
v1.0.0
Zvukogram 1.0.0 — Initial Release
- Добавлена генерация речи из текста через API Zvukogram с поддержкой SSML.
- Поддержка акцентирования, изменения скорости, пауз, псевдонимов для транскрипции английских слов.
- Возможность объединения аудиофрагментов (склейка) через ffmpeg.
- Предусмотрены готовые команды для проверки баланса и примеры вызова.
- Описаны варианты конфигурации, доступные голоса и ограничения API.
- Включены справочники по SSML, транскрипции, и раздел решения проблем.
元数据
常见问题
Zvukogram 是什么？
Text-to-Speech via Zvukogram API with SSML support. Use when you need to generate speech from text, create podcasts, voice notifications, or work with audio.... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 775 次。
如何安装 Zvukogram？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install zvukogram」即可一键安装，无需额外配置。
Zvukogram 是免费的吗？
是的，Zvukogram 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Zvukogram 支持哪些平台？
Zvukogram 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 Zvukogram？
由 erview（@erview）开发并维护，当前版本 v1.1.6。
推荐 Skills