ClawHub
Back to skill

Security audit

skill测试

Security checks across malware telemetry and agentic risk

Overview

This skill is mainly a read-only documentation evaluator, with a leftover dynamic-testing reference users should avoid treating as normal workflow.

Use this skill only for static review of specific Skill files you intend to evaluate. Do not let it run commands, use credentials, or follow the dynamic-testing reference unless you explicitly want a separate, sandboxed test workflow with clear user confirmation.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (4)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The example invocation includes a broad phrase like '帮我评估 weather skill 的质量', where verbs such as '评估/检查/测评' are common everyday wording and can cause the skill to trigger in situations beyond the author's intended scope. In an agent environment, overbroad triggers can lead to inappropriate routing, causing the evaluator skill to intercept unrelated requests and produce misleading analysis or interfere with other skills.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The skill advertises very broad trigger words such as “测评”, “评估”, “评测”, “文档检查”, “skill quality”, and “评分”, which are generic terms likely to appear in many unrelated user requests. In an agent environment, this can cause unintended invocation, misrouting, or overshadowing of more appropriate skills, reducing reliability and potentially exposing unrelated documents to unnecessary analysis.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
start_time = time.time()
    
    try:
        process = subprocess.Popen(
            command,
            shell=True,
            stdout=subprocess.PIPE,
Confidence
98% confidence
Finding
subprocess.Popen( command, shell=True

Tool Parameter Abuse

High
Category
Tool Misuse
Content
start_time = time.time()
    
    try:
        process = subprocess.Popen(
            command,
            shell=True,
            stdout=subprocess.PIPE,
Confidence
98% confidence
Finding
Popen( command, shell=True

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.