ClawHub
Back to skill

Security audit

胖叔 Skill 安全审查

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local security scanner for other skills, and its file-reading and optional install-hook behavior are consistent with that purpose.

Install this if you want a local heuristic scanner for skill directories. Keep automatic hooks scoped to incoming skill packages, scan explicit paths only, and treat clean results as useful signal rather than a guarantee.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
72% confidence
Finding
The skill metadata advertises automatic pre-install security vetting and explicitly references reading skill files and invoking a Python scanner, which implies file-read and shell-like execution capabilities. Having those capabilities without declared permissions weakens the trust boundary and can lead to the skill operating with more authority than reviewers or policy expect.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The description says the skill triggers automatically before installs and also when a user asks to 'vet' or 'check' a skill's security, which are broad everyday phrases that can cause over-invocation. Overly broad triggers increase the chance the skill runs in unintended contexts, potentially exposing files or launching scanning behavior when the user did not intend a security-sensitive action.

Static analysis

No suspicious patterns detected.