Security audit

FOSMVVM UI Tests Generator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Markdown-only generator for FOSMVVM SwiftUI UI test scaffolding, with some broad example templates but no hidden execution, persistence, credential handling, or exfiltration behavior.

Installers should treat this as a code-generation aid, not reviewed production architecture. Review generated SwiftUI view and operations templates before adopting them, especially any real API/data-loading implementation, and avoid keeping secrets in conversation or loaded code context used to generate tests.

A.I.G

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

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (5)

Description-Behavior Mismatch

Medium

Confidence: 90% confidence
Finding: The reference file for a UI test generator includes substantial production application scaffolding, including SwiftUI view implementations and operations code that performs business logic and data loading. This expands the skill's authority beyond test generation and can cause downstream agents or users to introduce unreviewed production code from a testing-focused skill, increasing the risk of unsafe logic, architectural drift, or accidental inclusion of network/business behavior.

Description-Behavior Mismatch

Medium

Confidence: 90% confidence
Finding: The operations template explicitly scaffolds real API-calling and asynchronous data-loading behavior inside a skill described as generating UI tests. This can mislead consumers into adopting security-relevant production code without appropriate review, especially because the sample normalizes embedding network/business logic in generated artifacts under the umbrella of test support.

Missing User Warnings

Low

Confidence: 92% confidence
Finding: The skill explicitly states it will reference prior conversation context and possibly view implementation already in context, but it does not clearly warn users that these materials become implicit inputs. This can lead to unintentional use of sensitive code or conversation data in generated outputs, especially in environments where users may not realize contextual data is being consumed automatically.

Intent-Code Divergence

Low

Confidence: 98% confidence
Finding: Earlier sections consistently state that display-only views do not need an operations property and should omit test-data transport. The checklist then directly contradicts that intent by requiring `operations` to be stored from `viewModel.operations` for views without operations, which would mislead users about what the skill actually expects them to generate.

Description-Behavior Mismatch

Low

Confidence: 77% confidence
Finding: Template 6 is presented as a view template and includes `AsyncImage(url:)`, which can trigger remote image fetching at runtime. For a skill whose stated purpose is generating UI tests, embedding production UI code with external-loading behavior is outside the described scope, even if only as a template.

Static analysis

No suspicious patterns detected.