ClawHub
Back to skill

Security audit

PIPL-Compliance

Security checks across malware telemetry and agentic risk

Overview

This is a local PIPL compliance and report-generation skill with disclosed file outputs and no evidence of hidden network, credential, persistence, or destructive runtime behavior.

Use this as a local drafting and self-check aid, not legal advice. Review output paths before running report or document generation because the tools write files locally, and consider pinning Jinja2 exactly if reproducible installs matter.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Intent-Code Divergence

Medium
Confidence
85% confidence
Finding
The document claims the skill will not modify files outside user-specified output paths, but the embedded example code deletes fixed filenames in the working directory unconditionally. If an agent or user follows this pattern, unrelated files with those names could be removed, creating integrity and availability risk and undermining the stated safety boundary.

Unpinned Dependencies

Low
Category
Supply Chain
Content
# 所有依赖版本锁定以确保运行稳定和通过安全扫描

pandas==2.2.0
jinja2>=3.1.6
Confidence
90% confidence
Finding
The dependency specifier `jinja2>=3.1.6` is not fully pinned, so future installs may resolve to different versions than were tested. While this does not introduce an immediate exploit by itself, it increases supply-chain and reproducibility risk because a later vulnerable or incompatible release could be pulled into the environment unexpectedly.

VirusTotal

55/55 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.