Security audit

FOSMVVM ViewModel Test Generator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent Markdown-only helper for generating FOSMVVM ViewModel test and localization template files, with no hidden execution or credential behavior found.

Before installing, confirm this skill matches your FOSMVVM Swift project and review generated test/YAML diffs, especially locale choices and inferred ViewModels. I found no evidence of hidden code execution, network access, credential handling, or persistence.

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

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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 (5)

Ae1

High

Category: analysis-evasion
Content: See [reference.md](reference.md) for complete file templates.
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected

Ae1

High

Category: analysis-evasion
Content: See [reference.md](reference.md) for complete file templates.
Confidence: 100% confidence
Finding: Referenced artifact was not completely inspected

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The skill explicitly states it will infer test structure from broad prior conversation and any code or YAML already in context, without defining boundaries on what context is authoritative or in-scope. In an agent setting, this can cause the model to incorporate stale, unrelated, or adversarially planted context into generated tests, producing incorrect files or leaking unintended project details into outputs.

Natural-Language Policy Violations

Low

Confidence: 88% confidence
Finding: The document states that translations are tested for all locales with 'en, es by default', which imposes a language/locale assumption in the skill behavior. Because no user choice or explicit opt-in is provided at that point, this may conflict with organizational language/locale policy expectations.

Natural-Language Policy Violations

Low

Confidence: 91% confidence
Finding: This markdown file states that testing uses additional locales beyond the default '(en, es)', and the examples throughout the file consistently assume English and Spanish. Under the policy rule for language/locale constraints, this is a natural-language preference that is presented as a default requirement without explicit user opt-in or justification.

Static analysis

No suspicious patterns detected.