Token导航 LogoToken导航

Agent Skill · 研究检索

openclaw-feishu-message(OpenClaw feishu message 搜索)

openclaw-feishu-message 用于查找、检索和筛选相关信息,适合在 OpenClaw 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

openclaw-feishu-message 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为只读;许可证为 MIT-0,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 31,062。

总安装 2,771周安装 111GitHub Stars 公开资料未说明下载 897核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
0.1.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:openclaw-feishu-message(OpenClaw feishu message 搜索)
来源仓库:https://github.com/systransform88/openclaw-feishu-message
安装命令:
openclaw skills install openclaw-feishu-message
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install openclaw-feishu-message

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

通过 OpenClaw 发送飞书私信与工作跟进消息。支持用户搜索、解析与定向创建个性化沟通内容。openclaw-feishu-message 属于研究检索类 Skill,可作为该场景下的辅助能力补充。适用于客户维护、内部通知与自动化提醒等用途。需配置正确的机器人身份与应用权限,否则消息无法送达。

群发消息需遵守飞书反垃圾规则,避免被限流或封号。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install openclaw-feishu-message;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

openclaw-feishu-message 怎么安装?

在终端执行 openclaw skills install openclaw-feishu-message。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

openclaw-feishu-message 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

openclaw-feishu-message 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

openclaw-feishu-message 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

openclaw-feishu-message 用的人多吗?

公开数据是累计安装 2,771 次,近一周安装 111 次,下载 897 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

openclaw-feishu-message 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
openclaw-feishu-message

OpenClaw feishu message 搜索 · 本页

2,771—只读MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
openclaw-feishu-message
description
Send Feishu/Lark direct messages and work follow-ups from OpenClaw. Use when the user wants to search for a person in Feishu/Lark, resolve a user, create a 1:1 chat, or send a bot-authored follow-up message about work done, status updates, reminders, or polite nudges.

Openclaw Feishu Message

Use this plugin to resolve a Feishu/Lark recipient and send a message as the configured bot.

Contact lookup trigger rules

When the user asks to search, find, look up, resolve, or identify a person in Feishu/Lark contacts, employee directory, or chat recipients, use feishu_message first.

This includes short requests like:

  • "search Ken"
  • "search Kai Lin"
  • "find Jennifer"
  • "look up WT Chim"
  • "search contact Kai Lin"
  • "find Jennifer in Lark"
  • "search employee by email"
  • "who is this Lark contact"
  • "find the open_id for SysTransform"

Do not claim there is no contact search tool when feishu_message is available in the runtime. Do not answer a contact search request from general reasoning alone when feishu_message can be called.

Core workflow

  1. Resolve the target.

- Prefer lookup_employee for exact identifiers: exact_name, email, or mobile. - Prefer find_contact for generic user prompts like search Ken, find Jennifer, or look up WT Chim. - Use search_employee for broader fuzzy queries or when you need a wider result set. - For generic prompts like search Ken, do not paraphrase from memory, call find_contact. - Prefer cached/open_id results over repeated fuzzy search when available.

  1. Cache the resolved contact.

- If a single match is found through lookup_employee or search_employee, let that tool path persist the contact cache. - For generic prompts like find Ken or search Ken, still route through search_employee or lookup_employee so cache updates. - Prefer tool-backed resolution over manual restatement so cache stays fresh.

  1. Return the resolved identity clearly.

- Include name, open_id, user_id, and email when present.

  1. Preview or compose the message.

- For work follow-up, prefer send_followup with dry_run=true first when ambiguity exists.

  1. Send only when the target is clear.

- For ordinary natural-language requests to message a person, resolve the person first, then send. - Prefer send_contact_message for ordinary natural-language requests to send a message to a person. - If the target is a person name like "SysTransform" or "Michael Chin", do not jump straight to raw-ID sending. Resolve/cache the person first, then send. - Use send_message when the target receive ID is already known. - For requests like "send a message to Michael" or "ask SysTransform for April PAS", attempt the real tool path first, not a drafted reply. - Do not invent chat-send actions or require chat creation when direct send can be used. - Use create_p2p_chat only if the user explicitly asked to create/open/start a chat. - Do not use create_p2p_chat as the default path for ordinary message sending. - Do not fall back to drafting text for the user unless the tool actually failed. - If the tool fails, report the real tool error briefly instead of a generic excuse. - If multiple people match the same name, stop and ask the user which one they mean.

Available actions

search_employee

Use to find likely employee matches from a name or keyword.

Example intent:

  • "Find John Tan in Lark"
  • "Search employee Chloe"

create_p2p_chat

Use only when the user explicitly asks to create/open/start a chat. Do not use create_p2p_chat for ordinary message requests when send_message can send directly.

send_message

Use when the target receive ID is already known and the exact text is ready.

send_contact_message

Use for ordinary requests like "send a message to Michael" or "tell SysTransform good night". This action should resolve the contact, cache the resolved identity, and send the message in one path. Prefer this over ad-libbed multi-step reasoning.

send_followup

Use for polite work-status nudges.

Recommended inputs:

  • target: one of open_id, user_id, union_id, email, chat_id, or name
  • work_item: short description of the completed work or task
  • status_prompt: optional custom ask
  • signature: optional sender sign-off
  • dry_run: use true first if the target was resolved by name only

Messaging guidelines

  • Keep follow-ups short and specific.
  • Mention the work item clearly.
  • Ask for one concrete update.
  • If the name lookup is fuzzy, do not send blindly.

Notes

  • This plugin depends on tenant scopes that allow employee lookup, chat creation, and bot messaging.
  • Tenant policy may still restrict who the bot can message even when scopes exist.

平台分布

OpenClaw

77.17%
按下载量换算692

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills