ClawHub
Back to skill

Security audit

Chart Craft Free

Security checks across malware telemetry and agentic risk

Overview

This is a local chart-generation skill with some confusing boilerplate about LLM/network use, but no artifact-backed evidence of hidden exfiltration, destructive behavior, or unsafe automatic actions.

Before installing, treat this as a local chart helper that will write chart data and generated images to your user directory. Avoid using highly sensitive datasets unless you are comfortable with local history files being retained, and clarify the publisher's intended networking model because the document mixes 'no network' claims with LLM API and callback_url references.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (4)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill repeatedly claims it is fully local and requires no network or external services, yet later introduces a required LLM API and network troubleshooting guidance. This creates a misleading trust boundary: users may provide sensitive data believing it never leaves the machine, when the documented dependencies imply external connectivity may occur.

Description-Behavior Mismatch

Medium
Confidence
91% confidence
Finding
The free-version description markets the tool as a simple local chart generator, but the dependency section states that an LLM API is required. That inconsistency can cause users to expose datasets to an external model provider under false assumptions of offline-only processing, especially harmful when chart inputs contain business or personal data.

Context-Inappropriate Capability

Low
Confidence
84% confidence
Finding
A callback_url parameter introduces outbound network interaction that is unnecessary for a local charting tool's stated purpose. Even if not directly exploited here, it expands the skill's capability surface and can enable unintended data exfiltration or SSRF-style behavior if later implemented naively.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The document describes local generation and persistent history storage but does not provide a clear upfront warning that inputs and chart metadata are written to disk and retained. Users handling sensitive data may unknowingly create recoverable local artifacts, increasing privacy and data retention risk on shared or managed systems.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.