← Back to Skills Marketplace
05u

久吾消息网关

by 05u · GitHub ↗ · v1.0.1 · MIT-0
cross-platform ✓ Security Clean
195
Downloads
1
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install jiuwu-message
Description
调用久吾消息网关HTTP接口给企业内部联系人发送消息。适用于:(1) 需要向企业内部同事发送通知或提醒时,(2) 需要批量发送内部消息时。当用户要求发送内部通知、提醒同事、向企业员工发消息、或调用消息网关时触发。
README (SKILL.md)

久吾消息网关

调用久吾消息网关发送企业内部消息。

环境配置

网关地址从环境变量 JIUWU_MESSAGE_GATEWAY_URL 读取,默认为 http://192.168.1.213:5000。可在 ~/.openclaw/workspace/.env~/.openclaw/.env 中配置。

脚本调用(优先使用)

始终使用 scripts/send_message.py,不要直接 HTTP 请求。

命令行调用

python scripts/send_message.py -c "工号" -t "消息内容" [-tt "标题"]

示例:

python scripts/send_message.py -c "1112" -t "您的报销单已审批通过" -tt "报销提醒"

批量发送

多个工号用英文逗号分隔:

python scripts/send_message.py -c "1112,1113,1114" -t "会议通知:今天下午2点开会" -tt "会议提醒"

Python 代码调用

import sys
sys.path.insert(0, 'skills/jiuwu-message/scripts')
from send_message import send_message

result = send_message(
    code="1112,1113",
    text="消息内容",
    title="可选标题"
)

参数说明

参数 缩写 必填 说明
--code -c 接收人工号,多个用英文逗号分隔
--text -t 消息内容
--title -tt 消息标题

响应格式

{
  "success": true,
  "data": true,
  "message": "请求成功"
}

失败时 successfalsemessage 包含错误信息。

Usage Guidance
This skill appears to do what it says: send messages to an internal gateway. Before installing, check the following: (1) Inspect any ~/.openclaw/workspace/.env and ~/.openclaw/.env files that the script will load—remove unrelated secrets or tokens you don't want the script to see. (2) Confirm the gateway URL (default is internal IP http://192.168.1.213:5000) and change it to the correct, secure endpoint (prefer HTTPS and authentication if available). (3) If your .env files contain credentials for other services, consider copying only the JIUWU_MESSAGE_GATEWAY_URL into a dedicated config for this skill to reduce exposure. (4) Review network policies—this skill will send outbound requests to the configured gateway. If you need stronger guarantees, add authentication to the gateway or restrict the agent's access to sensitive environment files.
Capability Analysis
Type: OpenClaw Skill Name: jiuwu-message Version: 1.0.1 The skill is a legitimate utility for sending internal messages via an HTTP gateway. The code in `scripts/send_message.py` uses standard Python libraries to perform POST requests to a configurable internal endpoint (defaulting to a private IP 192.168.1.213). It follows standard practices for configuration by reading environment variables from the OpenClaw workspace and contains no evidence of data exfiltration, malicious execution, or prompt injection.
Capability Assessment
Purpose & Capability
The name/description claim (call the 久吾 message gateway to send internal messages) matches the provided code and SKILL.md. The script constructs a JSON POST to /api/MessageGateway/SendMessagePost with the expected fields (code, text, optional title). No unrelated services, binaries, or credentials are requested.
Instruction Scope
SKILL.md confines runtime behavior to using scripts/send_message.py and documents the gateway URL env var. The script does load ~/.openclaw/workspace/.env and ~/.openclaw/.env to obtain configuration, which is related to the declared JIUWU_MESSAGE_GATEWAY_URL but broadens the data read at runtime (it parses all KEY=VALUE lines in those files). This is purposeful for configuration but grants the script access to any other env entries present in those files.
Install Mechanism
No install spec and no external downloads; the skill is instruction-plus-local Python script only. It uses the Python standard library (urllib) as claimed in requirements, so there is no extra-install risk.
Credentials
No required env vars are declared aside from optional JIUWU_MESSAGE_GATEWAY_URL (documented). However, the script explicitly loads and injects all variables from two .env files under ~/.openclaw, which could contain unrelated secrets or tokens for other tools if present. The skill does not request unrelated external credentials, but the loader increases exposure surface if those files exist and contain sensitive values.
Persistence & Privilege
The skill does not request persistent/always-on privileges, does not modify other skills or system-wide settings, and will only run when invoked. It does not write to system paths beyond reading the two .env files.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install jiuwu-message
  3. After installation, invoke the skill by name or use /jiuwu-message
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Added initial scripts and requirements files for the skill. - Clarified recommended usage: always use scripts/send_message.py, do not call HTTP endpoints directly. - Improved documentation for batch sending and typical usage scenarios. - Updated parameter descriptions for clarity. - Expanded and clarified response format documentation.
v1.0.0
- Initial release of jiuwu-message skill. - Enables sending messages to internal colleagues via the Jiuwu Message Gateway HTTP interface. - Supports specifying recipient employee code(s), message content, and an optional title. - CLI and Python code usage supported. - Reads gateway server URL from the `JIUWU_MESSAGE_GATEWAY_URL` environment variable.
Metadata
Slug jiuwu-message
Version 1.0.1
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is 久吾消息网关?

调用久吾消息网关HTTP接口给企业内部联系人发送消息。适用于:(1) 需要向企业内部同事发送通知或提醒时,(2) 需要批量发送内部消息时。当用户要求发送内部通知、提醒同事、向企业员工发消息、或调用消息网关时触发。 It is an AI Agent Skill for Claude Code / OpenClaw, with 195 downloads so far.

How do I install 久吾消息网关?

Run "/install jiuwu-message" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is 久吾消息网关 free?

Yes, 久吾消息网关 is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does 久吾消息网关 support?

久吾消息网关 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created 久吾消息网关?

It is built and maintained by 05u (@05u); the current version is v1.0.1.

💬 Comments