ClawHub
Back to skill

Security audit

Text Polish

Security checks across malware telemetry and agentic risk

Overview

This is a simple text-polishing skill made of markdown instructions, with no code, credentials, network access, or hidden persistence.

Safe to install based on the reviewed artifacts. Use explicit instructions when invoking it, and review the polished output before sending important text because broad rewrite requests can still change tone or wording.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
94% confidence
Finding
The skill’s trigger phrases are extremely broad and overlap with common everyday requests such as ‘检查这段文字’ or ‘改一改这段表达’. In an agent environment, this can cause unintended invocation on unrelated user content, increasing the chance of unwanted rewriting, privacy exposure of pasted text, or workflow interference when the user did not explicitly mean to call this skill.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The invocation guidance includes very generic phrases such as “润色”, “校对”, “改一改”, and “检查文字”, which are common in ordinary conversation and can cause the skill to activate when the user did not clearly intend to invoke it. Overly broad triggers increase the chance of unintended routing, accidental processing of unrelated content, and interference with other more appropriate skills.

Natural-Language Policy Violations

Medium
Confidence
85% confidence
Finding
The skill is defined as performing Chinese-language text polishing by default, without indicating user choice or fallback behavior for other languages. This can cause unintended transformation of multilingual or non-Chinese input, creating reliability and user-intent risks, especially if the agent routes non-Chinese text into a Chinese-only workflow.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal