ClawHub
Back to skill

Security audit

安全审计工具(免费版)

Security checks across malware telemetry and agentic risk

Overview

This is a local security-audit skill whose file reading, command use, and limited auto-fix behavior match its stated purpose, with some practical cautions around running auto-fix in the right project.

Install only if you want an agent to inspect the current project for secrets and security configuration issues. Run scans from the intended project directory, review findings before sharing reports because evidence may include secret fragments, and use auto-fix only when you are comfortable with local permission and .gitignore changes.

SkillSpector

Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
83% confidence
Finding
The skill advertises command execution and automatic remediation without sufficiently prominent warnings about user impact. In context, this is dangerous because the skill has exec capability and can change local filesystem state, so users may trigger scanning or remediation without understanding that commands will run or files will be modified.

Credential Access

High
Category
Privilege Escalation
Content
fixes = []
# ...
        # 修复.env权限
        env_file = Path(".env")
        if env_file.exists():
            os.chmod(env_file, 0o600)
            fixes.append("已设置.env文件权限为600")
Confidence
78% confidence
Finding
The auto-fix routine directly accesses and modifies a sensitive credential-bearing file (.env) in the current working directory without showing confirmation, backup, or target validation. In context, this is risky because the skill has exec-like operational authority and the fix path can affect sensitive files automatically, which may lead to unintended changes or disruption in multi-project or mis-scoped environments.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.