← Back to Skills Marketplace
richardcoder849

Multi-agent-bot-feishu

by Richardcoder849 · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ Security Clean
306
Downloads
1
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install multi-agent-bot-feishu
Description
在 OpenClaw 中创建新的 Agent 并绑定到新的飞书机器人/群聊。用于:(1) 添加新 Agent 管理不同飞书群聊 (2) 实现多机器人路由 (3) 为不同业务线创建独立 Agent。需修改 ~/.openclaw/openclaw.json 配置文件。
README (SKILL.md)

Multi-Agent-Bot 创建技能

此技能用于在 OpenClaw 中快速创建新 Agent 并配置飞书机器人绑定,实现多机器人多 Agent 架构。

使用场景

  • 为不同飞书群聊配置专属 Agent
  • 实现多机器人消息路由
  • 分离不同业务线的对话记忆
  • 隔离不同群聊的访问权限

工作原理

OpenClaw 支持多账户配置,通过 bindings 将不同的飞书机器人绑定到不同的 Agent,每个 Agent 有独立的工作空间和记忆。

创建步骤

步骤 1:准备信息

收集以下内容:

内容 说明 示例
Agent ID 唯一标识符 support, sales
Agent 名称 显示名称 "客服助手"
工作空间 独立目录路径 ~/.openclaw/workspace-support
飞书 App ID 机器人凭证 cli_xxx
飞书 App Secret 机器人密钥 xxx

步骤 2:修改配置文件

编辑 ~/.openclaw/openclaw.json,添加三部分配置:

1. 在 agents.list 添加新 Agent:

{
  "id": "新agent-id",
  "name": "显示名称",
  "workspace": "~/.openclaw/workspace-名称",
  "model": { "primary": "ark/doubao" }
}

2. 在 channels.feishu.accounts 添加机器人:

{
  "account-id": {
    "appId": "飞书appId",
    "appSecret": "飞书appSecret",
    "botName": "机器人名称",
    "dmPolicy": "allowlist",
    "allowFrom": ["允许的用户ID"]
  }
}

3. 在 bindings 添加路由:

{
  "agentId": "新agent-id",
  "match": {
    "channel": "feishu",
    "accountId": "account-id"
  }
}

完整配置模板见 references/config-template.json

步骤 3:创建工作空间

mkdir ~/.openclaw/workspace-名称

步骤 4:重启生效

openclaw gateway restart

路由优先级

当收到飞书消息时,按以下顺序匹配:

  1. 精确匹配peer.kind + peer.id(特定用户/群)
  2. 线程继承:群聊中线程消息
  3. 账户匹配accountId
  4. 渠道匹配channel + accountId: "*"
  5. 默认 Agent:设为 default: true 的 Agent

权限策略

策略 说明
open 允许所有人
allowlist 仅允许列表内用户
denylist 禁止列表内用户

注意事项

  1. 每个 Agent 必须有独立工作空间,避免记忆混淆
  2. 飞书机器人需先在飞书开放平台创建
  3. 修改配置后必须重启 OpenClaw
  4. 建议使用有意义的 Agent ID,便于识别

相关文档

Usage Guidance
This skill is an instruction-only guide to adding Agents and binding Feishu bots in OpenClaw and appears coherent. Before applying changes: back up ~/.openclaw/openclaw.json, verify you are editing the correct file, and restrict its filesystem permissions because the Feishu appSecret will be stored there in plaintext. Consider using a secret manager or environment-based secret injection if OpenClaw supports it. Test changes in a staging environment if possible and rotate app secrets if they are exposed. Finally, confirm the referenced OpenClaw docs link and that you trust the skill source before making config changes.
Capability Analysis
Type: OpenClaw Skill Name: multi-agent-bot-feishu Version: 1.0.1 The skill bundle is a configuration guide and template for setting up multi-agent Feishu (Lark) bot routing in OpenClaw. It contains no executable code, obfuscation, or malicious instructions; it purely provides documentation and a JSON template (references/config-template.json) to help users or agents manually configure the ~/.openclaw/openclaw.json file.
Capability Assessment
Purpose & Capability
Name/description match the instructions: the SKILL.md explains how to add agents, Feishu accounts, and bindings in ~/.openclaw/openclaw.json, create per-agent workspace directories, and restart the gateway. All requested configuration (appId/appSecret, agent id, workspace) is expected for this purpose.
Instruction Scope
Runtime instructions are limited to editing the specified OpenClaw config, creating workspace directories, and restarting the OpenClaw gateway. The skill does not instruct reading unrelated files, calling external endpoints, or exfiltrating data.
Install Mechanism
No install spec or code is included; the skill is purely documentation/instructions (lowest install risk).
Credentials
The skill does not request environment variables or extra credentials beyond what Feishu integration requires. It instructs the user to place Feishu appId/appSecret into ~/.openclaw/openclaw.json — this is proportionate to the function but means secrets will be stored in a plaintext config file; user should ensure proper file permissions or use a secret store if available.
Persistence & Privilege
always is false and the skill is user-invocable. It does not request permanent agent-wide privileges or modify other skills or system-wide settings beyond the user's OpenClaw config and workspaces (which is expected).
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install multi-agent-bot-feishu
  3. After installation, invoke the skill by name or use /multi-agent-bot-feishu
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Multi-Agent-Bot 在 OpenClaw 中快速创建新 Agent 并绑定飞书机器人,实现多机器人多 Agent 架构。 功能: • 为不同飞书群聊配置专属 Agent • 实现多机器人消息路由 • 分离不同业务线的对话记忆 • 隔离不同群聊的访问权限 使用方法: 1. 安装技能 2. 参考配置模板修改 openclaw.json 3. 创建独立工作空间 4. 重启 OpenClaw 需要 OpenClaw 3.10+ 和飞书官方插件
v1.0.0
## 1.0.0 (2026-03-19) ### 新增 - 初始版本发布 - 支持在 OpenClaw 中创建新 Agent 并绑定飞书机器人 - 提供完整配置模板和详细使用说明 - 支持多机器人多 Agent 架构配置
Metadata
Slug multi-agent-bot-feishu
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is Multi-agent-bot-feishu?

在 OpenClaw 中创建新的 Agent 并绑定到新的飞书机器人/群聊。用于:(1) 添加新 Agent 管理不同飞书群聊 (2) 实现多机器人路由 (3) 为不同业务线创建独立 Agent。需修改 ~/.openclaw/openclaw.json 配置文件。 It is an AI Agent Skill for Claude Code / OpenClaw, with 306 downloads so far.

How do I install Multi-agent-bot-feishu?

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

Is Multi-agent-bot-feishu free?

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

Which platforms does Multi-agent-bot-feishu support?

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

Who created Multi-agent-bot-feishu?

It is built and maintained by Richardcoder849 (@richardcoder849); the current version is v1.0.1.

💬 Comments