多个飞书

Automated setup and configuration of Feishu/Lark bots with OpenClaw. Use when creating multiple Feishu bots with independent workspaces, memories, and dedica...

Security Scan

The skill does what it claims (automates creating multiple Feishu/Lark bots), but the included script executes shell commands with untrusted config values and writes persistent credentials into your OpenClaw config—this introduces command-injection and secret-storage risks you should understand before running it.

Purpose & Capability

Name, description, SKILL.md, example config, and the Python script all align: the package automates creating OpenClaw agents, writing agent workspace files, adding Feishu account entries, and binding agents. The required Feishu app credentials are described in the docs and are used by the script.

Instruction Scope

The runtime instructions tell the agent/user to run the included Python script which: (a) creates agent workspaces and multiple files under ~/.openclaw/agents, (b) reads and updates ~/.openclaw/openclaw.json to add Feishu account credentials, (c) runs OpenClaw CLI commands via subprocess. These actions are in-scope for the stated goal but the script uses subprocess.run(shell=True) with string interpolation of values from the provided config (agentId, name, etc.), enabling command injection if the config is not fully trusted. The SKILL.md also refers to a fixed path (/root/.openclaw/...) for running the script which may not match a user's install location.

Install Mechanism

There is no install specification — the skill is instruction-only plus a bundled Python script. Nothing is downloaded or installed by the skill itself. Risk is limited to executing the included script and the OpenClaw CLI it calls.

Credentials

No environment variables or external credentials are required by the platform metadata, but the skill expects the user-supplied JSON config to include Feishu app credentials (appId, appSecret, encryptKey, verificationToken). Storing those secrets in ~/.openclaw/openclaw.json is expected for a channel integration but is persistent and should be considered sensitive. The skill does not request unrelated credentials.

Persistence & Privilege

The skill does not request always:true and will not be force-included. It writes files into the user's OpenClaw workspace and modifies the user's OpenClaw config (openclaw.json) — appropriate for setup tasks but a persistent change that affects system behavior and stores secrets.

What to consider before installing

This skill appears to implement multi-bot Feishu setup as advertised, but exercise caution before running the script: 1) Inspect the included scripts yourself — they will create agent folders and overwrite ~/.openclaw/openclaw.json (a backup is made). 2) Treat the JSON config as sensitive because it contains appSecret/encryptKey/verificationToken; do not paste secrets from untrusted sources. 3) The script constructs shell command strings using values from the config and calls subprocess.run(shell=True) — if any bot fields (agentId, name, etc.) contain attacker-controlled characters this can lead to command injection. Only run with configs you or trusted admins created. 4) Run the script in a controlled environment first (non-root account, staging instance, or VM). 5) Consider hardening the script (replace shell=True with argument lists, validate/sanitize agent names, and limit allowed characters) or ask the author to patch it before use. 6) After setup, rotate any secrets if you suspect exposure.

Feishu Bot Setup

Automated setup and configuration of Feishu/Lark bots with OpenClaw.

Features

• Flexible Role Definition - Customize role, style, and responsibilities for each bot
• Batch Configuration - Set up multiple bots with one command
• Independent Workspaces - Each bot has isolated storage and memory
• Auto-generated Configs - Automatically creates SOUL.md, IDENTITY.md, AGENTS.md
• WebSocket/Webhook - Support both connection modes

Overview

This skill automates the creation and configuration of multiple Feishu bots, each with:

• Independent agent workspace
• Dedicated memory and configuration files
• Feishu channel integration
• Proper routing and bindings

Prerequisites

• OpenClaw installed and running
• Gateway configured and accessible
• Feishu app credentials (App ID, App Secret, Encrypt Key, Verification Token)

Quick Start

1. Prepare Bot Configuration

Create a configuration file with bot details. All fields under personality are customizable:

{
  "bots": [
    {
      "name": "general-assistant",
      "agentId": "feishu-bot-1",
      "appId": "cli_xxx",
      "appSecret": "xxx",
      "encryptKey": "xxx",
      "verificationToken": "xxx",
      "connectionMode": "websocket",
      "personality": {
        "role": "通用助手",
        "tagline": "你的第一接待入口",
        "style": "温和简洁、快速响应",
        "styleDescription": "态度友好，回答简洁明了，快速响应",
        "responsibilities": [
          "常识问答",
          "轻量办公",
          "需求分流",
          "日常兜底"
        ],
        "description": "我是你的全能基础助理，快速响应你的各种需求。",
        "motto": "有问必答，有求必应，做你最贴心的第一入口。",
        "emoji": "🙋"
      }
    }
  ]
}

Personality Fields (All Optional, Customizable)

Field	Description	Example
role	Bot's role name	"数据分析助手"
tagline	Short tagline	"你的数据专家"
style	Working style	"严谨、高效"
styleDescription	Detailed style	"注重数据准确性，逻辑严密"
responsibilities	List of duties	["数据分析", "报表生成"]
description	Self-introduction	"我是数据分析专家..."
motto	Signature motto	"让数据说话"
emoji	Bot emoji	"📊"

You can define ANY role you need - customer service, sales, HR, finance, legal, etc.

2. Run Setup Script

python3 /root/.openclaw/workspace/skills/feishu-bot-setup/scripts/setup_bots.py /path/to/config.json

3. Restart Gateway

openclaw gateway restart

Manual Setup Steps

If you need more control, follow these manual steps:

Step 1: Create Agent

openclaw agents add <agent-id> --workspace /path/to/workspace --non-interactive

Step 2: Configure Feishu Channel

Edit ~/.openclaw/openclaw.json and add to channels.feishu.accounts:

"<bot-name>": {
  "appId": "cli_xxx",
  "appSecret": "xxx",
  "encryptKey": "xxx",
  "verificationToken": "xxx",
  "domain": "feishu",
  "connectionMode": "websocket",
  "webhookPath": "/webhook/feishu/<bot-name>",
  "dmPolicy": "open",
  "groupPolicy": "open",
  "requireMention": false,
  "reactionNotifications": "off",
  "typingIndicator": true,
  "resolveSenderNames": true
}

Step 3: Bind Agent to Feishu Account

openclaw agents bind --agent <agent-id> --bind feishu:<bot-name>

Step 4: Create Agent Configuration Files

Create these files in the agent workspace:

• SOUL.md - Role definition and personality
• IDENTITY.md - Identity information
• AGENTS.md - Workspace documentation
• TOOLS.md - Tool usage notes (optional)
• HEARTBEAT.md - Scheduled tasks (optional)

Step 5: Restart Gateway

openclaw gateway restart

Role Examples

You can create bots for ANY purpose. Here are some examples:

Customer Service Bot

{
  "role": "客服专员",
  "tagline": "你的贴心客服",
  "style": "耐心细致、专业友好",
  "responsibilities": ["问题解答", "投诉处理", "订单查询", "售后服务"],
  "emoji": "🎧"
}

Sales Assistant Bot

{
  "role": "销售顾问",
  "tagline": "你的销售专家",
  "style": "热情主动、洞察需求",
  "responsibilities": ["产品咨询", "方案推荐", "报价生成", "客户跟进"],
  "emoji": "💼"
}

HR Assistant Bot

{
  "role": "HR助手",
  "tagline": "你的人力资源伙伴",
  "style": "专业严谨、保密可靠",
  "responsibilities": ["政策解答", "请假审批", "入职指引", "福利咨询"],
  "emoji": "👔"
}

Legal Advisor Bot

{
  "role": "法务顾问",
  "tagline": "你的法律助手",
  "style": "严谨审慎、依法依规",
  "responsibilities": ["合同审核", "法律咨询", "风险提示", "合规建议"],
  "emoji": "⚖️"
}

Configuration Reference

Connection Modes

• websocket (default): Long-lived connection, no public URL needed
• webhook: Requires public URL for Feishu to send events

Policies

• dmPolicy: open, allowlist, pairing, disabled
• groupPolicy: open, allowlist, disabled
• requireMention: true or false

Troubleshooting

Bot Not Responding

1. Check gateway status: openclaw gateway status
2. Verify bindings: openclaw agents bindings
3. Check logs: openclaw logs --follow
4. Ensure Feishu app is published and events are subscribed

Credential Issues

• Verify App ID and App Secret match Feishu Open Platform
• Check Encrypt Key and Verification Token are correct
• Ensure app has required permissions (robot, im:message)

Resources

• scripts/setup_bots.py - Automated setup script
• references/example-config.json - Example configuration
• references/agent-templates/ - Agent configuration templates