data-lineage-tracker

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:41 PM UTC
• Version:
  1.0.0

Overview

This is a coherent construction data-lineage skill, but its optional AI analysis can send lineage details to SkillBoss API Hub.

Install only if your organization allows construction lineage metadata to be sent to SkillBoss API Hub. Use a dedicated API key, avoid including secrets or regulated data in lineage prompts, and scope filesystem access to the project files you intend to analyze.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Description-Behavior Mismatch

Medium

• Confidence: 94% confidence
• Finding: The skill’s stated purpose is lineage tracking, but this code exports full lineage context to a remote AI endpoint for analysis. Lineage data can contain sensitive metadata such as system names, owners, locations, transformation logic, and entity relationships, so sending it off-platform materially expands data exposure beyond what users would expect from a tracker.

Context-Inappropriate Capability

Medium

• Confidence: 88% confidence
• Finding: Remote LLM analysis is not necessary to perform basic lineage tracking, so its inclusion broadens the attack and privacy surface without clear functional need. This matters more in a compliance/audit context because the exported data may include sensitive provenance and transformation details that should remain internal.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The Quick Start directly instructs users to export complete lineage data and send it to an external API, normalizing insecure handling of potentially sensitive operational metadata. Example code is especially risky because users often copy it verbatim into production workflows.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The description mentions AI-powered analysis routing but does not clearly warn that lineage context may be sent to an external service. For a tool positioned around audit, compliance, and governance, lack of clear disclosure creates a significant transparency and privacy problem.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The Quick Start transmits exported lineage data to a third-party API without any adjacent warning or consent prompt. Because users commonly trust setup snippets, this omission can lead to unintentional disclosure of regulated or confidential project metadata.

VirusTotal

67/67 vendors flagged this skill as clean.