Feishu Voice Tts

将文本通过 MOSS-TTS 转换为语音，并发送到飞书群/个人。支持语音消息格式（带波形条）。

Security Scan

The code and SKILL.md implement Feishu TTS using MOSS-TTS and Feishu APIs, but the published metadata/requirements are inconsistent with the secrets and binary requirements the skill actually needs.

Purpose & Capability

The skill's stated purpose (convert text to speech via MOSS-TTS and send to Feishu) matches the scripts' behavior. However the registry metadata claims 'Required env vars: none' and 'Required binaries: none', while the code and SKILL.md clearly require MOSS_API_KEY, FEISHU_APP_ID, FEISHU_APP_SECRET and the ffmpeg binary. This mismatch is an incoherence between claimed requirements and actual needs.

Instruction Scope

The SKILL.md instructions are narrowly scoped and align with the code: call MOSS-TTS, transcode with ffmpeg, upload to Feishu, and optionally fetch message history. The scripts only access the Feishu and MOSS endpoints shown in the files. One minor red flag: SKILL.md's statement that 'Feishu configuration in OpenClaw was automatically completed' may be misleading — the scripts still require FEISHU_APP_ID/SECRET, and the metadata did not declare these.

Install Mechanism

There is no installer that downloads external code at runtime; the package is delivered as scripts in the skill bundle. No remote-install URLs or archive extraction steps were found. This is low install risk.

Credentials

The scripts legitimately require MOSS_API_KEY and FEISHU_APP_ID/FEISHU_APP_SECRET to operate; those environment variables are appropriate for the stated integration. The concern is that the skill metadata claims no required environment variables, which could lead users to inadvertently provide credentials without realizing. Required variables are named SECRET/KEY which is expected, but the metadata omission is a packaging inconsistency that affects user consent/awareness.

Persistence & Privilege

The skill does not request permanent platform presence (always: false). It does not modify other skills or system-wide settings. It runs as ad-hoc scripts and uses no elevated privileges beyond network calls to the noted APIs.

What to consider before installing

This skill's functionality appears to match its description (MOSS-TTS -> ffmpeg -> Feishu upload), but the published metadata omitted required environment variables and the requirement to install ffmpeg. Before installing or using it: 1) Do not supply secrets blindly — verify you want to give FEISHU_APP_ID/FEISHU_APP_SECRET and MOSS_API_KEY to this code and to the endpoints referenced (studio.mosi.cn and open.feishu.cn). 2) Confirm ffmpeg is installed in the environment. 3) Review the scripts (they are short and readable) to ensure they meet your policy — they call the listed external APIs and use subprocess/ffmpeg but do not contain obfuscated code or unknown endpoints. 4) Ensure the Feishu app permissions requested (message send, file upload, read) are acceptable and apply least privilege. 5) Ask the publisher to correct the skill metadata to declare required env vars and binaries; the current mismatch is misleading. If you cannot verify the MOSS endpoint or do not want to share these credentials, do not install/use the skill.

飞书语音 TTS 技能

当用户要求"给飞书发语音"、"在飞书里读这段话"或类似需求时，执行以下流程：

核心流程

1. MOSS-TTS 生成 WAV → 2. ffmpeg 转 opus → 3. 上传飞书 → 4. 发送语音消息

前置要求

• 系统已安装 ffmpeg
• 已配置环境变量 MOSS_API_KEY（MOSS-TTS API Key）
• 飞书配置在 OpenClaw 安装时已自动完成

使用方法

命令行模式

# 基本用法
python scripts/feishu_tts.py --text "要发送的文本" --chat_id "飞书群ID"

# 指定音色
python scripts/feishu_tts.py --text "文本" --voice_id "音色ID" --chat_id "群ID"

# 发送给个人（需要 open_id）
python scripts/feishu_tts.py --text "文本" --receive_id "用户open_id" --receive_id_type "open_id"

参数说明

参数	必填	说明
--text	✅	要转语音的文本
--chat_id	❌	飞书群 ID（与 receive_id 二选一）
--receive_id	❌	接收者 ID（open_id 或 chat_id）
--receive_id_type	❌	接收者类型：chat_id 或 open_id，默认 chat_id
--voice_id	❌	MOSS 音色 ID，默认 2001286865130360832（周周）
--output	❌	输出文件路径，默认 feishu_voice.wav

示例

# 给飞书群发语音
python scripts/feishu_tts.py --text "你好，我是 AI 助手" --chat_id "oc_xxx"

# 给指定用户发语音
python scripts/feishu_tts.py --text "你好" --receive_id "ou_xxx" --receive_id_type "open_id"

飞书权限要求

确保飞书应用已开通以下权限：

• im:message:send_as_bot - 以机器人身份发送消息
• drive:file:upload - 上传文件（发送语音消息必需）
• im:message.group:readonly - 读取群组消息（获取历史记录必需）
• im:chat:readonly - 读取会话列表

注意：需要在飞书开放平台应用配置中添加 drive:file:upload 权限并发布。

获取消息历史

脚本：get_history.py

获取飞书群消息历史，按时间排序。

# 获取最近 24 小时的消息
python scripts/get_history.py --chat_id "oc_xxx"

# 获取最近 1 小时的消息
python scripts/get_history.py --chat_id "oc_xxx" --hours 1

# 获取所有消息
python scripts/get_history.py --chat_id "oc_xxx" --all

# 只显示 audio 类型的消息
python scripts/get_history.py --chat_id "oc_xxx" --type audio

# 显示最近 10 条
python scripts/get_history.py --chat_id "oc_xxx" --limit 10

参数说明

参数	必填	说明
--chat_id	✅	飞书群 ID
--hours	❌	获取最近几小时的消息，默认 24
--all	❌	获取所有消息（不使用时间过滤）
--limit	❌	显示最近多少条消息，默认 30
--type	❌	只显示指定类型的消息，如 audio, file, text, image

环境变量

需要设置飞书应用认证：

FEISHU_APP_ID=你的应用ID
FEISHU_APP_SECRET=你的应用密钥