Join Tool

Security checks across malware telemetry and agentic risk

Overview

The skill is technically low-risk, but its script does not perform the SQL-style field join that it advertises.

Install only if you are comfortable treating this as a simple line-by-line concatenation helper, not a real join tool. Do not rely on it for CSV, security, accounting, or other decisions that require matching records by key unless the implementation or documentation is fixed.

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)

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding
The skill metadata and documentation claim SQL-like joining on a common field, but the analyzed behavior reportedly performs positional line concatenation and drops extra lines due to zip-like truncation. This can silently produce incorrect results, causing downstream automation or security decisions to be made on mismatched records while users believe they are getting key-based relational joins.

Description-Behavior Mismatch

Medium
Confidence
94% confidence
Finding
The implementation does not perform a join on a common field as advertised; it simply reads both files fully into memory and concatenates corresponding lines with zip(). This creates a semantic integrity issue: downstream users may trust the output as relationally joined data and make incorrect decisions, while unmatched or extra records are silently dropped due to zip truncation.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal