Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

CLI Anything Wrapper

v1.0.0

包装 CLI-Anything,让 OpenClaw 能调用任意软件的 CLI 功能

0· 63·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Suspicious
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description claim to wrap CLI-Anything to control local software; the code enumerates supported apps, locates harnesses under ~/.openclaw/workspace/CLI-Anything, and executes harness scripts. Required resources (none) align with this purpose.
Instruction Scope
SKILL.md and run.py stay within the wrapper's purpose, listing/installing CLI-Anything and invoking harness executables. Note: runtime will execute arbitrary local harness scripts with user-supplied arguments (via subprocess.run). This is expected for a wrapper but increases risk if untrusted inputs or harnesses are present.
Install Mechanism
There is no external install spec, but run.py includes an --install action that clones https://github.com/HKUDS/CLI-Anything (GitHub) and attempts ./setup.sh or pip install -e. Using a GitHub repo is standard, but the install runs remote setup scripts (shell=True), which can execute arbitrary code from the cloned repo — review upstream code before running install.
Credentials
The skill declares no environment variables or credentials and the code only reads the user's home path for the workspace. It does not request unrelated secrets or access tokens.
Persistence & Privilege
always:false and no modifications to other skills or system-wide settings. The skill can be invoked autonomously (default), which is normal for skills; combine that with its ability to run local binaries when considering risk.
Assessment
This skill is coherent with its description: it discovers and runs CLI-Anything harness scripts and can clone the CLI-Anything repo. Before installing or running it, review and trust the upstream CLI-Anything repo and any harness contents (they will be executed locally). Prefer dry-run mode first, avoid passing untrusted inputs as --args, and do not run --install unless you are comfortable executing a remote setup.sh/pip install from that GitHub repository. If you need stricter safety, run the wrapper in a sandbox or VM and inspect harness scripts before allowing autonomous agent use.

Like a lobster shell, security has layers — review code before you run it.

automationvk9797kf2ah407czxfj8pb2kwex84fztwclivk9797kf2ah407czxfj8pb2kwex84fztwlatestvk9797kf2ah407czxfj8pb2kwex84fztwsoftwarevk9797kf2ah407czxfj8pb2kwex84fztwwrappervk9797kf2ah407czxfj8pb2kwex84fztw

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments