← Back to Skills Marketplace
helong0911-alt

feishu-mention

by helong0911-alt · GitHub ↗ · v2.0.2 · MIT-0
cross-platform ⚠ suspicious
290
Downloads
0
Stars
1
Active Installs
3
Versions
Install in OpenClaw
/install feishu-mention
Description
**PRIORITY: HIGHEST**. **CRITICAL**: You **MUST** use this tool **BEFORE sending any message** to Feishu/Lark if it contains an "@" mention. **Trigger**: You...
README (SKILL.md)

Feishu Mention Resolver - 飞书@提及解析器

将飞书消息中的 @name 自动转换为 \x3Cat> XML 格式,实现更精确的提及提醒。

核心概念:Account ID

本工具使用 OpenClaw Account ID (accountId) 而非原始的飞书 App ID。 accountId 是在 ~/.openclaw/openclaw.json 中配置的账号别名(如 product, backend, elves)。

工具会自动读取配置文件,获取对应的 appIdappSecret,并自动发现配置中所有机器人的 OpenID。

快速开始

基本用法

// JavaScript/Node.js
const { resolve } = require('./index.js');

async function processMessage() {
  const text = '你好 @product,请确认一下';
  const accountId = 'elves'; // 使用 'elves' 账号的凭证来调用 API
  const chatId = 'oc_1234567890abcdef';
  
  // 解析文本中的所有@提及
  // 系统会自动识别 @product 是另一个机器人,并获取其 OpenID
  const resolvedText = await resolve(text, accountId, chatId);
  
  console.log(resolvedText);
  // 输出: "你好 \x3Cat user_id="ou_c610...">product\x3C/at>,请确认一下"
}

优先级逻辑

  1. 已配置的机器人 (Priority 1): 自动匹配 openclaw.json 中配置的所有账号名称(如 @product@Product@PRODUCT),不区分大小写。
  2. 用户别名 (Priority 2): 匹配 options.aliases 中的别名配置。
  3. 群成员 (Priority 3): 检查本地缓存或调用飞书 API 获取群成员。

API 参考

resolve(text, accountId, chatId, [options])

批量解析文本中的所有 @提及

参数:

  • text (string): 输入文本
  • accountId (string): OpenClaw 账号别名 (如 "elves")
  • chatId (string): 会话 ID
  • options (object): 可选配置
    • aliases: 额外的别名配置 [{ name: '张三', alias: ['小王'] }]

返回值:

  • Promise\x3Cstring>: 解析后的 XML 文本

版本: 2.0.0 (支持 Account ID 和自动发现)
作者: OpenClaw AI Assistant
最后更新: 2026-03-06

Usage Guidance
This skill appears to implement the advertised Feishu mention-resolution behavior, but it will read your OpenClaw configuration (~/.openclaw/openclaw.json) to obtain appId/appSecret for Feishu API calls and will write cache files under ~/.openclaw/workspace/cache/feishu_mentions. The registry metadata did NOT declare these config paths or credentials — double-check that you are comfortable allowing the skill to access those secrets. Before installing: (1) review the openclaw.json entries you have and ensure only intended accounts are present; (2) inspect the code (index.js and scripts) if possible to confirm network endpoints are only open.feishu.cn; (3) be aware caches will persist locally and may contain resolved OpenIDs; (4) consider running the skill in a restricted environment or sandbox if you cannot fully trust its source. If the missing required-config declaration is unexpected, ask the publisher to correct the metadata or provide a signed provenance for the package. If you want to change the behavior, you can restrict access by creating a config account with minimal permissions or removing appSecret from your global config and supplying credentials only at runtime.
Capability Analysis
Type: OpenClaw Skill Name: feishu-mention Version: 2.0.2 The feishu-mention skill is a utility designed to resolve @mentions in text into Feishu-compatible XML tags. It functions by reading credentials from the local OpenClaw configuration file (~/.openclaw/openclaw.json) to authenticate with the official Feishu API (open.feishu.cn) for retrieving bot and member IDs. The code (index.js and mention_resolver.py) implements standard caching mechanisms and follows the documented workflow without any signs of data exfiltration, unauthorized remote execution, or malicious prompt injection. All sensitive operations, such as reading API secrets, are directly tied to the skill's stated purpose of interacting with the Feishu platform.
Capability Assessment
Purpose & Capability
The code and docs implement a Feishu @mention resolver (converting @name to <at user_id=...>), which is coherent with the skill name and description. However, SKILL.md and the code require reading OpenClaw account configuration (appId/appSecret) from ~/.openclaw/openclaw.json to call Feishu APIs, while the registry metadata lists no required config paths or credentials — an inconsistency that should be explained.
Instruction Scope
Runtime instructions are narrowly scoped to parsing the draft message and resolving mentions by (1) checking configured bots, (2) aliases, (3) group members via Feishu API. The SKILL.md explicitly instructs the agent to obtain accountId and chatId from conversation context and to run feishu-mention.resolve before sending messages; this is consistent with the stated purpose. It does, however, require reading local config and using credentials to call Feishu APIs (documented in SKILL.md).
Install Mechanism
There is no install spec (instruction-only at registry level), but the package includes Node.js and Python code files that will read and write files when executed. Because nothing is formally installed by the skill metadata, users should note that running the code will create cache files under ~/.openclaw/workspace/cache/feishu_mentions and may perform network requests. The absence of an install declaration while shipping runnable code is a discrepancy to be aware of.
Credentials
The resolver needs access to Feishu credentials (appId/appSecret) stored in ~/.openclaw/openclaw.json in order to obtain tenant access tokens and fetch bot/member info. That credential requirement is proportionate to the stated function, but the top-level registry metadata does not declare required config paths or secrets. The skill will therefore read sensitive secrets from the user's OpenClaw config file and cache API results on disk — the mismatch between declared and actual required secrets/config paths is a notable concern.
Persistence & Privilege
The skill does not request always:true and does not modify other skills. It writes cache files to ~/.openclaw/workspace/cache/feishu_mentions (documented), and stores short-lived API tokens in memory. Writing a cache in the user's home directory is expected for this use case but is persistent and should be acceptable only if you trust the skill and its handling of cached data.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install feishu-mention
  3. After installation, invoke the skill by name or use /feishu-mention
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.0.2
- Initial publication of repository, including all source files and Git metadata - Updated SKILL.md to add `write` permission for the workspace cache path (`~/.openclaw/workspace/cache/feishu_mentions/**/*`) - No changes to user-facing features or APIs
v2.0.1
- Initial public release to version control. - Added all source, configuration, and documentation files. - Ready for integration and future development.
v2.0.0
Feishu Mention Resolver 2.0.0 introduces major changes: - Adds support for identifying accounts using OpenClaw Account ID instead of raw Feishu App ID. - Resolves all text "@mentions" to proper XML `<at>` tags for Feishu, ensuring users and bots are notified. - Automatically discovers all configured bot OpenIDs from local config. - Enforces correct workflow: always process message text with this tool before sending to Feishu if it contains "@mentions". - Improved prioritization for matching mentions (configured bots, aliases, then chat members).
Metadata
Slug feishu-mention
Version 2.0.2
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 3
Frequently Asked Questions

What is feishu-mention?

**PRIORITY: HIGHEST**. **CRITICAL**: You **MUST** use this tool **BEFORE sending any message** to Feishu/Lark if it contains an "@" mention. **Trigger**: You... It is an AI Agent Skill for Claude Code / OpenClaw, with 290 downloads so far.

How do I install feishu-mention?

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

Is feishu-mention free?

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

Which platforms does feishu-mention support?

feishu-mention is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created feishu-mention?

It is built and maintained by helong0911-alt (@helong0911-alt); the current version is v2.0.2.

💬 Comments