ClawHub
Back to skill

Security audit

Heatmap Beautifier

Security checks across malware telemetry and agentic risk

Overview

This appears to be a normal data-visualization skill with disclosed local file processing and ordinary Python plotting dependencies, with hygiene issues but no artifact-backed malicious behavior.

Before installing, use it only on CSV and annotation files you intend to share with the agent, and prefer installing in an isolated Python environment with pinned or reviewed dependency versions. The scan found dependency and documentation hygiene issues, but not hidden behavior, credential handling, persistence, or exfiltration.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (8)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill documentation instructs the agent to read user-supplied CSV files and local annotation files, which implies file-read capability, but the skill metadata declares no permissions. This mismatch can undermine sandboxing and trust decisions because an execution framework may allow undeclared file access or fail to present the true capability surface to users and reviewers.

Intent-Code Divergence

Medium
Confidence
93% confidence
Finding
The audit artifact is internally inconsistent: it asserts that exception-handling hardening was implemented, while later sections admit the underlying script may still retain a bare except and may not catch FileNotFoundError. This can cause downstream reviewers or deployment systems to trust a security/reliability improvement that was never actually verified, allowing weak error handling to persist unnoticed in production.

Unpinned Dependencies

Low
Category
Supply Chain
Content
matplotlib
numpy
pandas
seaborn
Confidence
97% confidence
Finding
The dependency is unpinned, so installs may resolve to different versions over time, reducing build reproducibility and increasing supply-chain risk if a newly released version is vulnerable or malicious. In a plotting/data-processing skill like this, that risk is real but indirect because the file does not itself force a known-bad version.

Unpinned Dependencies

Low
Category
Supply Chain
Content
matplotlib
numpy
pandas
seaborn
Confidence
99% confidence
Finding
The numpy dependency is unpinned, which allows environment-dependent resolution and can pull in versions with known security defects. Because the same file also flags numpy for multiple advisories, leaving it unpinned makes accidental installation of a vulnerable version more plausible.

Unpinned Dependencies

Low
Category
Supply Chain
Content
matplotlib
numpy
pandas
seaborn
Confidence
98% confidence
Finding
An unpinned pandas dependency creates non-reproducible environments and can expose consumers to vulnerable or incompatible releases. The risk is elevated somewhat because pandas also has a referenced advisory in this file, even though exploitability depends on how pandas features are used.

Unpinned Dependencies

Low
Category
Supply Chain
Content
matplotlib
numpy
pandas
seaborn
Confidence
96% confidence
Finding
The seaborn package is unpinned, so downstream installations may receive unexpected versions with security or compatibility regressions. This is a standard supply-chain hygiene issue rather than evidence of an actively malicious package declaration.

Known Vulnerable Dependency: numpy — 10 advisory(ies): CVE-2014-1859 (Numpy arbitrary file write via symlink attack); CVE-2021-41495 (NumPy NULL Pointer Dereference); CVE-2021-33430 (NumPy Buffer Overflow (Disputed)) +7 more

Critical
Category
Supply Chain
Confidence
84% confidence
Finding
The requirement allows installation of numpy versions associated with multiple published advisories, including memory-safety and file-write issues. Even if not every advisory is reachable in this heatmap skill, shipping an unconstrained dependency with known vulnerable versions materially increases the chance that users install an exploitable build.

Known Vulnerable Dependency: pandas — 1 advisory(ies): CVE-2020-13091 (** DISPUTED ** pandas through 1.0.3 can unserialize and execute commands from an)

High
Category
Supply Chain
Confidence
65% confidence
Finding
Pandas has a cited advisory related to unsafe deserialization/execution behavior, though the CVE is disputed and exploitability depends on the application invoking affected functionality on untrusted input. In this skill context, which processes data for visualization, that makes the issue less directly dangerous than a server-facing parser bug, but the unconstrained dependency can still expose users to risky versions.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.