← Back to Skills Marketplace
toto2016

Lark Mention

by toto2016 · GitHub ↗ · v26.3.27 · MIT-0
cross-platform ✓ Security Clean
127
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install lark-mention
Description
飞书 @ 提及专属技能。支持在群聊中向指定成员发送带真实 @ 提及的消息,解决飞书无法通过纯文本标签艾特成员的问题。触发场景:群里艾特成员、艾特全员、发送带 @ 的通知公告。
README (SKILL.md)

lark-mention — 飞书 @ 提及技能

核心能力

将自然语言转换为飞书标准 @ 提及消息,自动生成 mentions 字段并发送。

API 配置

LARK_BRIDGE_URL = http://localhost:18780/proactive

使用方式

方式一:通过 curl 调用

curl -X POST http://localhost:18780/proactive \
  -H "Content-Type: application/json" \
  -d '{
    "chatId": "\x3C群ID>",
    "text": "\x3Cat user_id=\"\x3Copen_id>\">成员名\x3C/at> 消息内容",
    "atOpenIds": ["\x3Copen_id>"]
  }'

方式二:通过 Node.js 模块调用

import { sendMention } from './scripts/lark-mention.mjs';

await sendMention({
  chatId: '\x3C群ID>',
  text: '请查收~',
  members: [
    { open_id: '\x3Copen_id>', name: '张三' },
    { open_id: '\x3Copen_id>', name: '李四' }
  ]
});

技术原理

飞书 @ 提及的正确格式

飞书消息 API 的 content 必须是 JSON 字符串,包含 textmentions 字段:

{
  "text": "\x3Cat user_id=\"\x3Copen_id>\">成员名\x3C/at> 消息内容",
  "mentions": [
    {
      "key": "\x3Copen_id>",
      "id": { "open_id": "\x3Copen_id>", "id_type": "open_id" },
      "name": "成员名"
    }
  ]
}

关键点:

  • text 中用 \x3Cat user_id="\x3Copen_id>">display_name\x3C/at> 占位
  • mentions 数组的 key 必须和 \x3Cat user_id="..."> 里的值完全一致
  • msg_type 必须是 "text"interactive 卡片类型不支持 mentions

常见错误

错误写法 原因
\x3Cat id="\x3Copen_id>"> 飞书不支持 id 属性,必须用 user_id
mentionskey 和 text 不匹配 导致渲染失败
msg_type 写成 interactive 卡片消息不支持 mentions 字段
纯文本写在 \x3Cat> 外部 飞书不渲染任何 \x3Cat> 标签

依赖

  • lark-openclaw-bridge 服务必须运行在 http://localhost:18780
  • 目标群聊需已添加飞书机器人
  • 使用者需提供:群ID、成员 open_id(从飞书后台获取)
Usage Guidance
This skill is coherent and low-risk if you run it in the intended environment: it expects a trusted local service 'lark-openclaw-bridge' at http://localhost:18780 to perform the actual Feishu API calls. Before installing, verify that the local bridge you use is trusted and does not forward data off-host or expose credentials. Note the module expects you to supply chatId and members' open_id values (obtained from Feishu); the skill itself does not request or store API keys. Also ensure your runtime provides fetch (or polyfill) if you import the .mjs module.
Capability Analysis
Type: OpenClaw Skill Name: lark-mention Version: 26.3.27 The lark-mention skill is a utility designed to format and send messages with @mentions to Lark (Feishu) group chats via a local bridge service. The implementation in scripts/lark-mention.mjs is straightforward, constructing the required JSON payload and sending it to a local endpoint (http://localhost:18780/proactive). No evidence of data exfiltration, malicious execution, or prompt injection was found.
Capability Assessment
Purpose & Capability
Name/description (Feishu @ mentions) match the included code and SKILL.md. Required inputs (chatId, members' open_id and names) are exactly what you'd expect for constructing mentions; no unrelated credentials or services are requested.
Instruction Scope
SKILL.md and README instruct POSTing to a local bridge (http://localhost:18780/proactive) or importing the provided module. Instructions do not ask the agent to read unrelated files, environment variables, or send data to external endpoints beyond the documented local bridge.
Install Mechanism
There is no install spec (instruction-only skill with a small JS module). No downloads from external URLs or archive extraction; risk from installation is minimal.
Credentials
The skill declares no required environment variables, credentials, or config paths. The only network dependency is a local bridge service; this is proportionate to the stated purpose (the bridge presumably handles Feishu auth).
Persistence & Privilege
always is false and the skill does not request persistent system privileges or modify other skills' configuration. Autonomous invocation is allowed (platform default) and not excessive here.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install lark-mention
  3. After installation, invoke the skill by name or use /lark-mention
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v26.3.27
Initial release: real @mention in Lark groups
v1.0.0
Initial release: real @mention in Lark groups
Metadata
Slug lark-mention
Version 26.3.27
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is Lark Mention?

飞书 @ 提及专属技能。支持在群聊中向指定成员发送带真实 @ 提及的消息,解决飞书无法通过纯文本标签艾特成员的问题。触发场景:群里艾特成员、艾特全员、发送带 @ 的通知公告。 It is an AI Agent Skill for Claude Code / OpenClaw, with 127 downloads so far.

How do I install Lark Mention?

Run "/install lark-mention" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Lark Mention free?

Yes, Lark Mention is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Lark Mention support?

Lark Mention is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Lark Mention?

It is built and maintained by toto2016 (@toto2016); the current version is v26.3.27.

💬 Comments