ClawHub
Back to skill

Security audit

Agent项目必备文件管理器

Security checks across malware telemetry and agentic risk

Overview

This skill transparently creates and maintains local project governance files to help agents track rules, versions, and handoffs.

Install this only for projects where you want persistent governance files. Review the generated AGENTS.md before relying on it, customize the permission zones and autonomy levels, and avoid --force unless you intentionally want to replace existing governance files.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill instructs the agent to run shell commands and create or modify multiple files, but it does not declare permissions or clearly scope those capabilities. That creates a trust-boundary problem: users or hosting systems may not realize the skill can write to the project and execute commands, increasing the chance of unintended file changes or command execution in the wrong directory.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The skill's behavior includes scaffolding governance files and updating them every session, but the description does not prominently warn that it will create and modify multiple files in the target project. This can lead to surprising bulk changes, accidental overwrites, or repository pollution if the user invokes it without understanding its side effects.

Natural-Language Policy Violations

Medium
Confidence
92% confidence
Finding
The template includes Chinese explanatory text in the primary onboarding section, which can steer the agent toward a specific language/locale without explicit user opt-in. In a governance skill that is read every session, persistent language bias can reduce usability, cause mismatched outputs, and confuse collaborators expecting a different language, though it is not directly a security compromise.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.