1227cwx/telegram-bot-chat/SKILL.md
# Telegram Bot间群聊通信技能 **创建时间**: 2026-02-20 11:03 UTC+8 **技能版本**: 1.0.0 **适用场景**: Telegram群聊中多agent协同通信 name: telegram-bot-chat version: 1.0.0 description: Telegram群聊中实现多agent之间互相通信的技术,促进协同工作效率提升 homepage: https://telegram.org metadata: {"category":"communication","platform":"telegram","use_case":"multi_agent_coordination"} # Telegram Bot群聊通信 实现多agent在Telegram群聊中的高效协同通信,促进协同工作效率的提升 ## Skill Files | File | URL | |------|-----| | **SKILL.md** (this file) | Current file | | **SKILL.md** | Skill
npx skillsauth add openclaw/skills 1227cwx/telegram-bot-chatInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
创建时间: 2026-02-20 11:03 UTC+8 技能版本: 1.0.0 适用场景: Telegram群聊中多agent协同通信
name: telegram-bot-chat version: 1.0.0 description: Telegram群聊中实现多agent之间互相通信的技术,促进协同工作效率提升 homepage: https://telegram.org metadata: {"category":"communication","platform":"telegram","use_case":"multi_agent_coordination"}
实现多agent在Telegram群聊中的高效协同通信,促进协同工作效率的提升
| File | URL | |------|-----| | SKILL.md (this file) | Current file | | SKILL.md | Skill documentation |
Install locally:
# 安装到工作空间skills目录
cp -r telegram-bot-chat /root/.openclaw/workspace/skills/
然后主agent需要手动执行多bot部署:
# 主agent执行:复制skill到每个bot的工作空间
WORKSPACE_SKILLS="/root/.openclaw/workspace/skills"
# 复制workspace中所有skills到每个bot的workspace目录
for skill_dir in $WORKSPACE_SKILLS/*/; do
if [ -d "$skill_dir" ]; then
skill_name=$(basename "$skill_dir")
echo "复制技能 $skill_name 到每个bot的工作空间"
# 复制到workspace-2, workspace-3, workspace-4等
for workspace_dir in /root/.openclaw/workspace*; do
if [ "$workspace_dir" != "$WORKSPACE_SKILLS" ]; then
mkdir -p "$workspace_dir/skills"
cp -r "$skill_dir" "$workspace_dir/skills/"
echo "已部署技能 $skill_name 到 $workspace_dir"
fi
done
fi
done
Or just read from this file!
⚠️ IMPORTANT:
🔒 关键要点:
适用性检查: 所有agent都可使用此技能进行多agent协同通信
Telegram Bot API不会把bot发的消息投递给其他bot,所以:
当需要与其他bot对话时,使用:
sessions_send(
sessionKey="agent:bot2:telegram:group:群ID", // 正确的sessionKey
message="群聊对话。来自:main。内容:想要bot2回复的内容..."
)
当收到其他bot通过sessions_send发来的消息时:
message(
action="send",
channel="telegram",
accountId="自己的bot accountId", // 使用自己的bot accountId
target="群ID", // 群ID
message="你的回复内容"
)
agent:[bot名称]:telegram:group:群ID
示例:
agent:bot2:telegram:group:群IDagent:bot3:telegram:group:群IDagent:bot4:telegram:group:群IDbotA → sessions_send → botB收到 → botB回复 → OpenClaw announce → 群聊显示
accountId:使用自己的bot accountIdchannel:固定为"telegram"target:群ID每个bot必须在channels.telegram.accounts中有完整配置:
// 错误的做法
// @bot3 在群里(bot3收不到)
// 正确的做法
sessions_send(
sessionKey="agent:bot3:telegram:group:群ID",
message="bot2在群里说:bot3,过来帮忙!"
)
// bot3收到sessions_send后
message(
action="send",
channel="telegram",
accountId="自己的bot accountId", // 使用自己的bot accountId
target="群ID",
message="来了,什么事?"
)
// 错误的sessionKey
agent:bot2:main // 没有channel和群信息
// 正确的sessionKey
agent:bot2:telegram:group:群ID
// 错误:没有accountId
message(action="send", channel="telegram", target="群ID")
// 正确:包含accountId
message(action="send", channel="telegram", accountId="自己的bot accountId", target="群ID")
第1步:安装到主agent工作空间
# 安装到工作空间skills目录
cp -r telegram-bot-chat /root/.openclaw/workspace/skills/
第2步:主agent手动执行多bot部署
# 复制skill到每个bot的工作空间(主agent执行)
WORKSPACE_SKILLS="/root/.openclaw/workspace/skills"
# 复制workspace中所有skills到每个bot的workspace目录
for skill_dir in $WORKSPACE_SKILLS/*/; do
if [ -d "$skill_dir" ]; then
skill_name=$(basename "$skill_dir")
echo "复制技能 $skill_name 到每个bot的工作空间"
# 复制到workspace-2, workspace-3, workspace-4等
for workspace_dir in /root/.openclaw/workspace*; do
if [ "$workspace_dir" != "$WORKSPACE_SKILLS" ]; then
mkdir -p "$workspace_dir/skills"
cp -r "$skill_dir" "$workspace_dir/skills/"
echo "已部署技能 $skill_name 到 $workspace_dir"
fi
done
fi
done
第3步:验证部署
多bot部署说明:
通过这个技能,多个agent可以在Telegram群聊中实现高效协同通信,促进协同工作效率的提升。关键是要使用正确的sessionKey格式和message工具参数。
检查更新: 随时重新阅读此文件查看新功能!
tools
Use when the user wants to connect to, test, or use the McDonalds service at mcp.mcd.cn, including checking authentication, probing MCP endpoints, listing tools, or calling McDonalds MCP tools through a reusable local CLI.
development
Web scraping platform — Twitter/X data, Vinted marketplace, and general web scraping API
development
SlowMist AI Agent Security Review — comprehensive security framework for skills, repositories, URLs, on-chain addresses, and products (Claude Code version)
data-ai
去除中文文本中的 AI 写作痕迹,使其读起来自然。基于维基百科 AI 写作特征指南,检测 24 种 AI 模式。触发词:humanizer-cn、去除 AI 痕迹、去除 AI 写作痕迹、中文文本人性化。