Back to skill

Security audit

Homestruk Deal Analyzer

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward real-estate deal analysis skill, with a limited local report-save instruction that users should be aware of.

Before installing, be aware that completed analyses may be saved locally and can include property addresses, prices, financing assumptions, and cash-flow estimates. Use it for investment modeling, but review outputs independently before making financial decisions.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
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)

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill instructs the agent to write a deal analysis file into a local workspace path, which is a side effect beyond the core analytical behavior described in the manifest. Even though the target path appears limited and business-related, automatic persistence can expose sensitive financial/property data, create unwanted files, and surprise users if performed without explicit consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
96% confidence
Finding

The documentation describes saving output to disk without any warning, confirmation, or user-controlled choice. Silent file creation is risky because users may not expect local persistence of potentially sensitive investment assumptions, addresses, and cash-flow analysis, and an agent could perform this side effect automatically.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.