← 返回 Skills 市场
219
总下载
0
收藏
0
当前安装
2
版本数
在 OpenClaw 中安装
/install meeting-minutes-qa-tts
功能描述
Read meeting minutes, produce a short summary with the current conversation model, save the meeting text and summary into local memory, answer follow-up ques...
安全使用建议
This skill appears to do what it says: summarize meeting text, store it locally, and generate mp3s via SenseAudio. Before installing or running it, consider the following: (1) the skill will read any local path or URL you provide — do not point it at sensitive files (e.g., private keys, system configs); (2) the SenseAudio API key (SENSEAUDIO_API_KEY) or a key you type will be sent to api.senseaudio.cn along with the text to produce audio — avoid sending confidential meeting text unless you trust that service and your account; (3) the runtime needs Python and the 'requests' library available (the skill does not include an install spec), so ensure your environment satisfies that; (4) the skill will write mp3 files to the output path you choose and will create directories as needed — avoid paths that could overwrite important files. If you want greater assurance, review the generate_meeting_audio.py and related scripts yourself or run them in an isolated environment.
功能分析
Type: OpenClaw Skill
Name: meeting-minutes-qa-tts
Version: 1.0.1
The skill bundle provides legitimate functionality for summarizing meeting minutes and converting them to audio using the SenseAudio TTS API. The code in `scripts/generate_meeting_audio.py` and `scripts/read_meeting_text.py` facilitates reading from local files or URLs and writing MP3 files to user-specified paths, which are necessary for the stated purpose. While the ability to read arbitrary files or fetch URLs carries inherent risk in an agentic workflow, the implementation lacks any signs of malicious intent, obfuscation, or unauthorized data exfiltration. The instructions in `SKILL.md` are transparent and align with the provided Python logic.
能力评估
Purpose & Capability
Name and description match the code and SKILL.md: the scripts read meeting text (local file or URL), save a JSON meeting memory under memory/, and call the SenseAudio TTS API (api.senseaudio.cn) to produce mp3 output. No unrelated cloud credentials, system-wide config paths, or extra services are requested.
Instruction Scope
SKILL.md narrowly instructs the agent to ask for a meeting text or path/URL and an output path, to prefer an existing SENSEAUDIO_API_KEY, and to keep memory local by default. The scripts follow this. Be aware the skill will read arbitrary local file paths or fetch arbitrary URLs if provided — this is expected for a 'read meeting note' skill but means the agent can be made (by user input) to read any file the runtime user can access.
Install Mechanism
This is instruction/code-only: no install spec is provided. The code uses the third-party 'requests' library but the skill does not declare any dependency metadata — the runtime must provide Python and requests. No remote archives or external installers are fetched by the skill itself.
Credentials
The only primary credential used is a SenseAudio API key (SENSEAUDIO_API_KEY), which is proportionate to TTS functionality. Minor inconsistency: generate_meeting_audio also reads optional env vars SENSEAUDIO_DEFAULT_VOICE and SENSEAUDIO_TTS_MAX_CHARS (seen in code) but those are not documented in SKILL.md's requires.env; SKILL.md documents SENSEAUDIO_API_KEY correctly.
Persistence & Privilege
The skill writes local JSON memory under its own memory/ directory by default and writes mp3 files to user-specified paths; always:false and it does not request permanent platform-level privileges or alter other skills. This is within expected scope.
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 meeting-minutes-qa-tts - 安装完成后，直接呼叫该 Skill 的名称或使用
/meeting-minutes-qa-tts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
v1.0.1
- Now checks if SENSEAUDIO_API_KEY is present in the environment before prompting the user for a SenseAudio API key.
- Standardizes usage of SENSEAUDIO_API_KEY across all Python-related calls in the skill.
- Added several test and cache files to improve coverage and stability.
- Removed unused files, including README.md and a temporary meeting text file.
v1.0.0
- Initial release of "meeting-minutes-qa-tts" skill.
- Reads a meeting note, generates a short summary using the current conversation model, and saves both text and summary locally.
- Answers follow-up questions about the same meeting, providing both text and audio (mp3) outputs.
- Integrates with the SenseAudio TTS API for speech synthesis.
- Prompts users for meeting source, SenseAudio API key, and mp3 output paths as needed.
- All audio files and meeting memories are managed locally within the skill directory by default.
元数据
常见问题
meeting-minutes-qa-tts 是什么？
Read meeting minutes, produce a short summary with the current conversation model, save the meeting text and summary into local memory, answer follow-up ques... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 219 次。
如何安装 meeting-minutes-qa-tts？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install meeting-minutes-qa-tts」即可一键安装，无需额外配置。
meeting-minutes-qa-tts 是免费的吗？
是的，meeting-minutes-qa-tts 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
meeting-minutes-qa-tts 支持哪些平台？
meeting-minutes-qa-tts 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 meeting-minutes-qa-tts？
由 曹广雨（@xiaocaijic）开发并维护，当前版本 v1.0.1。
推荐 Skills