Openclaw Installer Cn

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: openclaw-installer-cn Version: 1.0.0 The skill provides automated installation and diagnostic capabilities for OpenClaw, specifically optimized for users in China. It includes high-risk instructions such as using 'sudo' to modify system file permissions and executing remote scripts via 'curl | bash' (specifically the NVM installer from raw.githubusercontent.com). While these actions are aligned with the stated purpose of environment setup and troubleshooting, they involve significant system-level changes and potential RCE risks if the agent executes them without user oversight. No evidence of intentional data exfiltration or malicious backdoors was found in SKILL.md.

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

A user asking for a repair could have packages, permissions, and configuration changed before reviewing the exact commands.

Why it was flagged

The skill tells the agent that a quick-fix request should automatically perform mutating repair actions, but it does not define approval gates or precise limits for those actions.

Skill content
### 快速修复

```
修复 OpenClaw 安装问题
```

自动执行:
- 清理缓存
- 重装依赖
- 修复权限
- 更新配置
Recommendation

Make diagnosis the default, show the exact commands and target paths, and require explicit user approval for each mutating repair step.

What this means

This could weaken or break system package ownership and affect other Node/npm tools on the machine.

Why it was flagged

These commands require elevated privileges and recursively change ownership of npm and global Node module paths, including a protected system-wide location.

Skill content
sudo chown -R $(whoami) ~/.npm
sudo chown -R $(whoami) /usr/local/lib/node_modules
Recommendation

Prefer a user-local npm prefix or nvm; if a permission fix is still needed, ask for explicit approval and scope it to the minimum verified OpenClaw-related path.

ConcernHigh Confidence
ASI05: Unexpected Code Execution
What this means

If the remote source or delivery path is compromised, arbitrary commands could run on the user's machine.

Why it was flagged

The skill recommends piping a downloaded remote script directly into a shell, with no checksum, signature verification, or inspection step.

Skill content
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.39.0/install.sh | bash
Recommendation

Download and inspect installer scripts first, verify integrity from official documentation, and require user confirmation before execution.

ConcernHigh Confidence
ASI08: Cascading Failures
What this means

If run in the wrong project, this could remove a lockfile, change dependency resolution, and break or alter unrelated work.

Why it was flagged

The cleanup commands delete dependency artifacts in the current directory without specifying that the directory has been verified as an OpenClaw install path.

Skill content
npm cache clean --force
rm -rf node_modules
rm package-lock.json

# 重装
npm install
Recommendation

Validate the target directory, back up lockfiles, and ask the user before any rm or reinstall command.

What this means

The user may install whatever package version is served by the mirror at that time.

Why it was flagged

Global package installation from an external mirror is relevant to the installer purpose, but the package versions and provenance checks are not specified.

Skill content
npm install -g cnpm --registry=https://registry.npmmirror.com
cnpm install -g openclaw
Recommendation

Use official sources where possible, pin known-good versions, and verify package provenance before global installation.

What this means

API keys placed in shell profiles can be exposed if dotfiles are shared, backed up insecurely, or copied into support requests.

Why it was flagged

The skill proposes storing provider API keys as persistent shell-profile environment variables; this is expected for model-provider setup, but it is not declared in the registry metadata.

Skill content
export DEEPSEEK_API_KEY="your-key-here"
export ZHIPU_API_KEY="your-key-here"
Recommendation

Enter real API keys manually, avoid sharing shell profiles, and consider a secrets manager or restricted-permission env file.