ClawHub
Back to skill

Security audit

html-report

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed HTML report-generation skill with no hidden persistence, credential access, or destructive behavior found.

Before installing, be aware that the skill may create local HTML output files, use a headless browser for screenshots, load external CDNs or fonts in generated HTML, and search the web for unfamiliar diagram structures. Review generated HTML before sharing it if the report contains private or sensitive information.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The template explicitly instructs the agent to perform network/WebSearch behavior when a requested diagram is not already covered by the local library. That expands the skill from local HTML/SVG report generation into external data retrieval, which can violate expected capability boundaries, create privacy leakage if user prompts are sent to search, and introduce prompt-injection or untrusted-content risks from fetched results.

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
This section defines a complete WebSearch workflow for discovering diagram structures from the internet, despite the skill being described as an HTML report generator. Such instruction broadens the trust boundary to external sources and can expose the system to malicious search results, data exfiltration through queries, and nondeterministic behavior not implied by the skill's stated purpose.

Natural-Language Policy Violations

Medium
Confidence
94% confidence
Finding
The workflow content is entirely written in Chinese and prescribes fixed output conventions without indicating that language should follow user preference or an explicit parameter. In an agent skill, this can override or ignore a user's requested language, causing incorrect outputs, degraded usability, and downstream misunderstandings when reports are consumed by users expecting another language.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The file is authored entirely in Chinese and presents mandatory guidance without any indication that language should be selected based on user preference. In an agent skill, forcing a specific output language can override user intent, reduce transparency, and cause misunderstandings in downstream tasks or security-relevant reports.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.