← 返回 Skills 市场
67
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install bot-roundtable
功能描述
Enables multiple Bots in a Feishu group to spawn expert personas and hold realistic multi-Bot discussions automatically or on demand.
使用说明 (SKILL.md)
Bot Roundtable Discussion
多Bot圆桌讨论系统 - 让飞书群里的多个Bot像真人一样讨论问题。
功能
- 单个Bot收到问题，自动spawn多个专业分脑
- 各分脑通过不同Bot身份发送到群聊
- 群里呈现多Bot讨论效果，像真人群聊
触发词
- "讨论 XXX"
- "大家怎么看"
- "多角度分析"
- "辩论 XXX"
使用方法
自动触发
在飞书群里说"讨论 XXX"即可自动触发
手动调用
python skills/bot-roundtable/coordinator.py "讨论问题"
手动spawn
from feishu_sender import send_message_as_bot
# spawn分脑
sessions_spawn(label="tech", task="技术专家分析：XXX")
sessions_spawn(label="product", task="产品专家分析：XXX")
# 发送结果
send_message_as_bot("qiwang", chat_id, "技术观点...")
send_message_as_bot("meidou2", chat_id, "产品观点...")
文件结构
bot-roundtable/
├── SKILL.md # 本文件
├── feishu_sender.py # 飞书多Bot发消息模块
├── coordinator.py # 协调脚本
├── config.json # Bot凭证配置
└── README.md # 详细文档
配置
config.json:
{
"bots": {
"default": {"app_id": "...", "app_secret": "..."},
"qiwang": {"app_id": "...", "app_secret": "..."},
"meidou2": {"app_id": "...", "app_secret": "..."}
},
"group_chat_id": "oc_xxx"
}
依赖
- Python 3.8+
- 飞书开放平台应用（需配置 app_id + app_secret）
注意事项
- 每条消息间隔0.5秒，避免频率限制
- 消息300字以内
- Bot凭证从 openclaw.json 的 channels.feishu.accounts 读取
安全使用建议
Do not install this skill yet. Ask the author for: (1) the missing Python files (coordinator.py, feishu_sender.py) and config.json so you can inspect their logic; (2) a clear list of every credential or config file the skill will read (it currently references openclaw.json but declares nothing); (3) confirmation of where openclaw.json lives and what other secrets it contains. Only proceed if the code is available for review and the skill limits itself to only the Feishu bot credentials it needs (and documents that). If you must test, run it in a tightly sandboxed environment with temporary Feishu test accounts (not production credentials).
功能分析
Type: OpenClaw Skill
Name: bot-roundtable
Version: 1.0.0
The skill bundle describes a multi-bot orchestration system for Feishu (Lark) group chats, allowing an agent to spawn multiple 'expert' personas to discuss topics. The provided files (SKILL.md, package.json, and _meta.json) outline legitimate functionality aligned with the stated purpose, and although the system handles sensitive credentials (app_id/app_secret), there is no evidence of intentional data exfiltration, malicious execution, or harmful prompt injection.
能力评估
Purpose & Capability
The stated purpose (spawn multiple Feishu Bots) legitimately requires Feishu app_id/app_secret values and a config path; however the skill metadata declares no required environment variables or credentials. SKILL.md references config.json and also says Bot credentials are read from openclaw.json channels.feishu.accounts — a platform config path that was not declared. This lack of declared credentials is incoherent with the described functionality.
Instruction Scope
Runtime instructions tell the agent to run Python scripts (coordinator.py, feishu_sender.py), read config.json and openclaw.json for Bot credentials, and send messages as multiple bot identities. Those files are not actually included in the package. Instructions also direct reading the agent/platform config (openclaw.json), which is outside the skill's declared surface and could expose other accounts — this broad file access is a red flag.
Install Mechanism
There is no install spec (instruction-only), which is lower risk for arbitrary code installation. However package.json lists files (feishu_sender.py, coordinator.py, config.json) that are not present in the published bundle; that mismatch reduces confidence and suggests the SKILL.md was written for a different package layout.
Credentials
The skill declares no required env vars or primary credential, yet the instructions require app_id/app_secret for multiple bots and explicitly say bot credentials are read from openclaw.json. Requesting access to platform/agent config without declaring it is disproportionate and could enable exfiltration of unrelated credentials stored there.
Persistence & Privilege
always:false (no forced permanence) and normal autonomous invocation settings. Autonomous invocation combined with the skill's implied access to platform credentials (openclaw.json) increases risk, but persistence/privilege flags themselves are not elevated.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install bot-roundtable - 安装完成后，直接呼叫该 Skill 的名称或使用
/bot-roundtable触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
- Initial release of Bot Roundtable Discussion skill.
- Enables multiple bots to simulate a group discussion in Feishu chats, appearing as different expert perspectives.
- Supports automatic and manual discussion triggers with customizable bots for flexible analysis and debate.
- Includes configuration file for managing multiple bot credentials.
- Requires Python 3.8+ and Feishu Open Platform apps.
元数据
常见问题
Bot Roundtable 是什么？
Enables multiple Bots in a Feishu group to spawn expert personas and hold realistic multi-Bot discussions automatically or on demand. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 67 次。
如何安装 Bot Roundtable？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install bot-roundtable」即可一键安装，无需额外配置。
Bot Roundtable 是免费的吗？
是的，Bot Roundtable 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Bot Roundtable 支持哪些平台？
Bot Roundtable 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Bot Roundtable？
由 linbo405（@linbo405）开发并维护，当前版本 v1.0.0。
推荐 Skills