← Back to Skills Marketplace
99
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install wecom-gui-message
Description
通过 macOS GUI 自动化给企业微信（WeCom）联系人或群发消息。使用 peekaboo + screencapture + Swift Vision OCR + cliclick 实现全流程自动化。适用于需要通过企业微信桌面端发送消息的场景，如推送日报、通知等。触发词：企微发消息、企业微信发消息、wec...
Usage Guidance
This skill appears internally consistent with its stated purpose, but consider the following before installing:
- Security/privacy: It requires macOS Screen Recording and Accessibility permissions so it can read your screen and control the UI — while functioning it will capture screenshots (including any visible sensitive information) and store them under /tmp/wecom-gui. Only run it when you are comfortable with that and avoid having sensitive windows visible.
- Metadata mismatch: The SKILL.md and code require helper tools (peekaboo, cliclick, swift) but the declared registry requirements list none. Verify you install these tools from trusted sources and that the author updates the metadata.
- Review binaries: peekaboo and cliclick will be executing UI actions; prefer official releases or package manager installations and inspect their sources if possible.
- Least privilege: Run in a controlled environment (test account or VM) first, and do not grant permissions unless you need the automation. If you plan to allow autonomous invocation, be aware an agent with this skill can control the WeCom desktop UI while it runs.
Capability Analysis
Type: OpenClaw Skill
Name: wecom-gui-message
Version: 3.0.1
The skill performs GUI automation for Enterprise WeChat (WeCom) on macOS, requiring high-privilege permissions including Screen Recording and Accessibility. It uses `subprocess.run(shell=True)` in `scripts/send_message.py` to execute system commands and interact with third-party tools like `cliclick` and `peekaboo`. While the logic is consistent with the stated purpose of sending messages via OCR and coordinate-based clicking, the use of shell execution with user-provided message strings (despite basic quote escaping) poses a potential shell injection risk. No evidence of intentional data exfiltration or malicious intent was found.
Capability Assessment
Purpose & Capability
The name/description (WeCom GUI message sender) matches the code and instructions: the Python driver + Swift OCR + use of cliclick/peekaboo/screencapture implement exactly that functionality. However, the registry metadata lists no required binaries/env vars while SKILL.md and the code clearly require external tools (peekaboo, cliclick, swift). This metadata omission is an inconsistency and should be corrected but does not by itself indicate malicious intent.
Instruction Scope
SKILL.md and scripts restrict actions to local macOS GUI automation: activating the WeCom app, window resizing, window screenshots, local OCR, clicking and pasting into the app. The instructions do not read unrelated files or environment variables, nor do they call external network endpoints. Important privacy note: the skill takes screenshots and runs OCR on them (including QR codes and any visible text), so it can capture any visible sensitive information on the screen while running.
Install Mechanism
This is an instruction-only skill with no install spec or remote downloads; that reduces risk because no additional code is being fetched during install. The runtime does call local tools (swift, peekaboo, cliclick, screencapture), so those must be installed separately from trusted sources.
Credentials
The skill requests no environment variables or credentials and only needs system permissions (Screen Recording and Accessibility) appropriate to GUI automation. The lack of credentials is proportional to the stated purpose.
Persistence & Privilege
always:false (not force-enabled) and default model invocation settings are used. The skill does not request persistent system-wide changes nor attempt to modify other skills' configs. It will create temporary files under /tmp/wecom-gui; this is expected for its operation.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install wecom-gui-message - After installation, invoke the skill by name or use
/wecom-gui-message - Provide required inputs per the skill's parameter spec and get structured output
Version History
v3.0.1
fix: remove personal names from code examples
v3.0.0
v3.0: 窗口截图替代全屏截图、坐标偏移换算、关闭右侧面板、发送前验证
Metadata
Frequently Asked Questions
What is wecom-gui-message?
通过 macOS GUI 自动化给企业微信（WeCom）联系人或群发消息。使用 peekaboo + screencapture + Swift Vision OCR + cliclick 实现全流程自动化。适用于需要通过企业微信桌面端发送消息的场景，如推送日报、通知等。触发词：企微发消息、企业微信发消息、wec... It is an AI Agent Skill for Claude Code / OpenClaw, with 99 downloads so far.
How do I install wecom-gui-message?
Run "/install wecom-gui-message" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is wecom-gui-message free?
Yes, wecom-gui-message is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does wecom-gui-message support?
wecom-gui-message is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created wecom-gui-message?
It is built and maintained by jacky-wzj (@jacky-wzj); the current version is v3.0.1.
More Skills