Security audit

Real Estate Report Workflow

Security checks across malware telemetry and agentic risk

Overview

This is a coherent real-estate report workflow, but users should run it in a dedicated project folder because it can search the web, create files, and run report-processing scripts.

Install only if you want an AI-assisted real-estate report workflow that can perform web research, create local Markdown/DOCX files, install dependencies, and run bundled Python/Node scripts. Use a dedicated folder, review output paths before export, avoid sensitive client data in shared workspaces, and approve any Git commit or dependency install explicitly.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Missing User Warnings

Low

Confidence: 89% confidence
Finding: The README explicitly documents network-enabled tools (`web_search` / `web_fetch`) and states that an automatic environment self-check runs on first use, but it does not warn users that external requests may occur or describe what data might be transmitted. In an agent skill context, missing disclosure can lead to unintended outbound access, privacy surprises, or execution in environments where network/tool use is not expected.

Vague Triggers

Medium

Confidence: 92% confidence
Finding: The trigger description includes broad phrases such as '工作流', '市场调查报告', '竞品分析', 'SWOT分析', and '运营预测', which can match many unrelated user requests. This can cause accidental invocation of a tool-heavy workflow that instructs environment checks, file reads, web access, and script execution, increasing the chance of unnecessary data access or disruptive behavior in the wrong context.

Missing User Warnings

Low

Confidence: 84% confidence
Finding: The script writes merged content to INPUT_MD, which is directly influenced by the REPORT_MD environment variable, without constraining the destination to an expected workspace. In an agent or automation context, an attacker who can influence environment variables could cause arbitrary file overwrite within the executing user's permissions, potentially clobbering application files or sensitive local data.

Unpinned Dependencies

Low

Category: Supply Chain
Content: python-docx>=1.0.0
Confidence: 94% confidence
Finding: python-docx>=1.0.0

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.