Feishu Send Message

Send private or group messages via the current Agent's Feishu app, showing the configured bot name and handling app-isolated open_id.

Security Scan

The skill's code and instructions align with its stated purpose: it reads the agent's local Feishu app credentials and uses the official Feishu API to send private/group messages; nothing in the files suggests hidden endpoints or unexplained privileges.

Purpose & Capability

Name/description (Feishu message sending) match the implementation: the script reads the agent-specific config (~/.openclaw/openclaw-{agent}.json) for appId/appSecret and calls open.feishu.cn to obtain a token and post messages. Required tools (curl/jq) listed in metadata are appropriate.

Instruction Scope

Instructions and the script operate within the expected scope (read agent config, call Feishu API). The SKILL.md also instructs grepping system journal logs (journalctl) to find user open_id; this is functionally necessary to obtain open_id but means the skill will access local service logs and could surface additional log contents. The script echoes App ID, target and message to stdout (may appear in logs) — not strictly required and may leak non-secret context.

Install Mechanism

No install spec; this is an instruction-only skill with a small bash script included. No external downloads or archive extraction are performed.

Credentials

The skill does not request environment variables but reads local config files that contain the Feishu app_id and app_secret. Access to those files is proportionate for sending messages, but those files contain sensitive credentials (app_secret). The metadata expects jq/curl; ensure those are trusted binaries on PATH.

Persistence & Privilege

always is false and the skill does not persist new credentials or modify other skills. It runs on-demand and does not request elevated system privileges.

Assessment

This skill appears to do what it says: it reads the agent's local Feishu config (~/.openclaw/openclaw-{agent}.json) to obtain app_id and app_secret and posts to the official Feishu API. Before installing, verify you trust the skill source (homepage/source unknown), confirm the config files contain only expected credentials, and be aware the SKILL prints the App ID, target and message (which may end up in shell or agent logs). If you do not want the skill to read system logs to extract open_id, manually obtain open_id and call the script. Reject the skill if you see network targets other than open.feishu.cn or any obfuscated/remote-downloaded code in future versions.

feishu-send-message Skill

飞书多 Agent 消息发送技能。让 Ass/Ops 用自己的飞书应用发送消息（群聊/私聊），解决 open_id 应用隔离问题。

核心功能

• 使用当前 Agent 的飞书凭证发送消息
• 支持群聊和私聊两种模式
• 飞书显示对应的机器人名字（如"小助手"或"运维助手"）

使用场景

• Boss 派发任务后，Ass/Ops 用自己的身份回复消息
• 需要显示不同机器人名字的场景

使用方法

方法 1：直接使用脚本

# 获取用户 open_id（从当前 Agent 网关日志）
journalctl --user -u openclaw-gateway-ass.service | grep "received message from"

# Ass 发送私聊消息
/home/admin/.openclaw/workspace-ass/skills/feishu-send-message/send.sh \
 ass <open_id> open_id "消息内容"

# Ass 发送群聊消息
/home/admin/.openclaw/workspace-ass/skills/feishu-send-message/send.sh \
 ass oc_88fbbe55e37eca3c3339b2f4bae1a8a9 chat_id "群聊消息"

# Ops 发送私聊消息
/home/admin/.openclaw/workspace-ops/skills/feishu-send-message/send.sh \
 ops <open_id> open_id "运维消息"

方法 2：直接使用 curl 命令

# 读取当前 Agent 配置（以 Ass 为例）
APP_ID=$(jq -r '.channels.feishu.appId' ~/.openclaw/openclaw-ass.json)
APP_SECRET=$(jq -r '.channels.feishu.appSecret' ~/.openclaw/openclaw-ass.json)

# 获取 token
TOKEN=$(curl -s -X POST https://open.feishu.cn/open-apis/auth/v3/tenant_access_token/internal/ \
  -H 'Content-Type: application/json' \
  -d "{\"app_id\":\"$APP_ID\",\"app_secret\":\"$APP_SECRET\"}" | jq -r '.tenant_access_token')

# 发送私聊消息
curl -s -X POST "https://open.feishu.cn/open-apis/im/v1/messages?receive_id_type=open_id" \
  -H "Authorization: Bearer $TOKEN" \
  -H 'Content-Type: application/json' \
  -d "{\"receive_id\":\"ou_xxx\",\"msg_type\":\"text\",\"content\":\"{\\\"text\\\":\\\"消息内容\\\"}\"}"

# 发送群聊消息
curl -s -X POST "https://open.feishu.cn/open-apis/im/v1/messages?receive_id_type=chat_id" \
  -H "Authorization: Bearer $TOKEN" \
  -H 'Content-Type: application/json' \
  -d "{\"receive_id\":\"oc_xxx\",\"msg_type\":\"text\",\"content\":\"{\\\"text\\\":\\\"群聊消息\\\"}\"}"

参数说明

参数	说明	示例
agent	Agent 类型：ass 或 ops	ass
target	目标 ID（open_id 或 chat_id）	<open_id> 或 <chat_id>
msg_type	消息类型：open_id（私聊）或 chat_id（群聊）	open_id
message	消息内容	你好，这是测试消息

关键参数参考

参数	Ass 值	Ops 值	说明
App ID	cli_a924c897d4b89ceb	cli_a924d0d16bf99cee	当前 Agent 应用的 ID
App Secret	从配置文件读取	从配置文件读取	当前 Agent 应用的密钥
用户 open_id	从日志获取	从日志获取	当前 Agent 应用中的用户 ID
群聊 chat_id	oc_88fbbe55e37eca3c3339b2f4bae1a8a9	同左	群聊 ID（应用通用）

获取用户 open_id

从当前 Agent 网关日志获取：

# Ass
journalctl --user -u openclaw-gateway-ass.service | grep "received message from"

# Ops
journalctl --user -u openclaw-gateway-ops.service | grep "received message from"

注意事项

1. open_id 是应用隔离的：

   • 每个应用有独立的用户 ID 体系
   • 必须使用当前 Agent 应用中的 open_id 发送私聊消息

2. chat_id 是应用通用的：

   • 同一个群聊，不同应用看到的是同一个 chat_id
   • 可以直接使用

3. content 格式：

   • 必须是转义的 JSON 字符串："{\"text\":\"消息内容\"}"
   • 不是嵌套对象

4. 配置文件位置：

   • Ass: ~/.openclaw/openclaw-ass.json
   • Ops: ~/.openclaw/openclaw-ops.json
   • Boss: ~/.openclaw/openclaw-boss.json

自动化脚本

脚本位置：/home/admin/.openclaw/workspace-{agent}/skills/feishu-send-message/send.sh

#!/bin/bash
# 参数：$1=agent (ass/ops), $2=target (open_id 或 chat_id), $3=msg_type (open_id/chat_id), $4=message

AGENT=$1
TARGET=$2
MSG_TYPE=$3
MESSAGE=$4

# 读取配置
APP_ID=$(jq -r ".channels.feishu.appId" ~/.openclaw/openclaw-${AGENT}.json)
APP_SECRET=$(jq -r ".channels.feishu.appSecret" ~/.openclaw/openclaw-${AGENT}.json)

# 获取 token
TOKEN=$(curl -s -X POST https://open.feishu.cn/open-apis/auth/v3/tenant_access_token/internal/ \
  -H 'Content-Type: application/json' \
  -d "{\"app_id\":\"$APP_ID\",\"app_secret\":\"$APP_SECRET\"}" | jq -r '.tenant_access_token')

# 发送消息
curl -s -X POST "https://open.feishu.cn/open-apis/im/v1/messages?receive_id_type=${MSG_TYPE}" \
  -H "Authorization: Bearer $TOKEN" \
  -H 'Content-Type: application/json' \
  -d "{\"receive_id\":\"$TARGET\",\"msg_type\":\"text\",\"content\":\"{\\\"text\\\":\\\"$MESSAGE\\\"}\"}" | jq '.data.message_id'