Insight Brief Generator

• Outcome: Pass
• Latest audit: May 28, 2026, 1:31 PM
• Version:
  1.0.0

Overview

This skill is a local insight-brief generator; it has some unrelated dormant audit code, but the shipped configuration only runs the disclosed briefing workflow.

Install this as a local briefing tool, not a security scanner. Provide only files you intend it to read, choose the output path carefully because it can create or overwrite that file, and review generated briefs before using them for business decisions.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 88% confidence
• Finding: The skill declares no permissions, yet its instructions explicitly allow running `python3` and imply reading local resources and writing output files. This creates a capability/permission mismatch that can mislead users and orchestration systems about what the skill is able to do, weakening policy enforcement and review controls.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 93% confidence
• Finding: A skill presented as a management-insight summarizer is reported to perform directory scanning, content inspection, pattern-based secret/risk discovery, and dynamic audit modes. That behavioral mismatch is dangerous because it can cause users or calling agents to supply the skill in low-risk contexts while it actually processes unrelated local files and potentially sensitive repository content.

Description-Behavior Mismatch

High

• Confidence: 97% confidence
• Finding: The script supports multiple operational modes such as directory audit, CSV audit, pattern audit, and skill audit, which materially exceed the declared purpose of generating management-readable insight briefs from reports and charts. This scope expansion increases access to local files and enables broad inspection behaviors that are unnecessary for the stated skill, raising the risk of unintended data exposure and misuse.

Context-Inappropriate Capability

High

• Confidence: 98% confidence
• Finding: The code performs security-style pattern scanning for secrets, private URLs, and dangerous shell commands, then reports matched snippets from local files. In a skill presented as an executive insight brief generator, this behavior is unjustified and creates a clear capability to inspect arbitrary content for sensitive material, which could expose confidential data during routine use.

Context-Inappropriate Capability

Medium

• Confidence: 92% confidence
• Finding: The script recursively enumerates files in directories and summarizes their contents, including filenames, extensions, and Markdown headings. For a briefing skill, filesystem-wide auditing is out of scope and can reveal internal project structure or document content that users did not intend to include in an executive summary workflow.

VirusTotal

VirusTotal findings are pending for this skill version.