ClawHub

Chart Generator

Security checks across malware telemetry and agentic risk

Overview

This is a local chart-generation skill with some documentation inconsistencies, but its file reads and document outputs are disclosed in the body and fit the chart/reporting purpose.

Install in a virtual environment when possible, review the generated Python before running it, and choose input/output paths deliberately. Treat generated Word, HTML, and Markdown files as shareable documents that may contain embedded chart data, and avoid using untrusted titles or captions in HTML/Markdown without escaping.

SkillSpector

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

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The manifest materially understates the skill's capabilities versus the documented implementation: it advertises only a narrow set of chart types and PNG/SVG output, while the code supports additional data ingestion modes and export targets including PDF, Word, HTML, Markdown, base64 embedding, and directory aggregation. This mismatch weakens policy enforcement and user/operator trust because reviewers may approve or invoke the skill under false assumptions about what it can access and produce.

Intent-Code Divergence

Medium
Confidence
91% confidence
Finding
The security notes claim 'Local file processing only,' but the feature list explicitly includes 'web' as a data source. Contradictory security claims are dangerous because they can cause reviewers and users to underestimate data exposure paths or permit the skill in environments where network-enabled or externally sourced content is restricted.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal