Real Estate Showing Brief

• Outcome: Pass
• Latest audit: May 28, 2026, 11:59 AM
• Version:
  1.0.0

Overview

This skill is a local real-estate showing brief generator with disclosed Python/file use, though its helper script contains dormant generic audit code that users should avoid repurposing.

Install only if you want a local drafting helper for real-estate showing preparation. Use it with intended buyer/property input files, choose output paths carefully, avoid feeding unnecessary personal or sensitive client data, and do not repurpose the bundled script as a general directory or secret scanner.

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 (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 90% confidence
• Finding: The skill declares no explicit permissions, yet its instructions include use of python3 plus local file inputs/outputs, which implies shell execution and file read/write capability. This creates a transparency and least-privilege problem: users and the platform may treat the skill as low-risk while it can access or transform local data and invoke an interpreter.

Description-Behavior Mismatch

High

• Confidence: 98% confidence
• Finding: The script dispatches to multiple generic audit modes such as directory, CSV, pattern, and skill-package auditing that are unrelated to a real-estate showing brief skill. This materially expands the skill's capability surface so a user or upstream system could repurpose it to inspect local files and generate reports about arbitrary directories, which is dangerous in a constrained skill context.

Context-Inappropriate Capability

High

• Confidence: 97% confidence
• Finding: The code contains a generic regex-based scanner for secrets, internal URLs, and risky shell patterns over arbitrary files or directories. In a real-estate brief skill, this is out of scope and enables unintended local content inspection, which could expose sensitive material from unrelated files even if the output masks only part of some matches.

Context-Inappropriate Capability

Medium

• Confidence: 95% confidence
• Finding: The skill_audit routine performs generic package structure and frontmatter compliance checks on arbitrary skill directories, which is unrelated to preparing a buyer/showing brief. This broadens the tool into a repository inspection utility and can reveal file presence, metadata structure, and package contents from local directories outside the declared use case.

VirusTotal

VirusTotal findings are pending for this skill version.