ClawHub
Back to skill

Security audit

党政机关公文格式标准排版

Security checks across malware telemetry and agentic risk

Overview

This is a local document-formatting skill that reads user-provided documents, writes Word/report outputs, and shows no hidden upload, credential use, or persistence behavior.

Install only if you want local processing of official-document drafts. Provide only the files you intend it to process, review the style-cleaning report before relying on wording changes, and use dry-run/review-only when exact language matters.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill explicitly instructs the agent to use terminal/Bash and local scripts, read user-provided files, write output .docx and reports, and inspect the local font environment, yet no permissions are declared. This creates a capability/permission mismatch that can lead to overbroad file or environment access at runtime and weakens policy enforcement and user expectations about what the skill is allowed to do.

Natural-Language Policy Violations

Medium
Confidence
83% confidence
Finding
The skill defines a default path where, if the user only provides text without specifying an action, it automatically runs style_clean in mode C and rewrites the text into a government-style register. Because this transformation happens without explicit opt-in, the agent may alter user content, tone, or meaning unexpectedly, which is risky for official documents where wording precision matters.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.