Security audit

Frontend Premium

Security checks for vulnerabilities and agentic risk

Overview

This is an opinionated frontend design helper that may add design/checklist files, but it contains only markdown guidance and no hidden code, network access, or credential handling.

Install this only if you want an opinionated frontend workflow that may create a design.md and push modern UI/accessibility conventions by default. For small UI fixes, tell the agent to skip the full design-system workflow if that is not desired.

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

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The skill metadata describes broad frontend assistance without clear activation boundaries, which can cause the agent to apply prescriptive behavior to unrelated or only partially related tasks. In an agentic system, ambiguous auto-activation increases the chance of unintended instruction injection into workflows, file creation, or design constraints the user did not request.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The phrase 'When working on frontend tasks' is overly broad and makes the core workflow apply to an undefined set of situations. Because the workflow is framed as enforced, an agent may generate extra artifacts like design.md or alter execution order even when the task is minor, unrelated, or constrained by user requirements.

Vague Triggers

Medium

Confidence: 96% confidence
Finding: Saying the skill will 'automatically provide guidance' for frontend tasks implies implicit activation without a precise scope or user consent checkpoint. This can override task intent by injecting mandatory process steps, technology choices, and file generation into broad categories of work, creating reliability and prompt-scope risks.

Natural-Language Policy Violations

Low

Confidence: 87% confidence
Finding: The phrase 'RSC优先' introduces Chinese-language content into the instructions while the rest of the file is in English, and there is no indication that users opted into this locale choice. This can conflict with language consistency expectations and may force a language/locale preference implicitly.

Natural-Language Policy Violations

Low

Confidence: 76% confidence
Finding: The markdown mandates 'Accessibility (WCAG 2.1 AA)' as a fixed requirement in natural language, without indicating whether this standard is optional, user-selected, or tied to a specific compliance context. Under the stated policy, prescriptive locale/standard requirements can be problematic when they are imposed without opt-in or justification.

Static analysis

No suspicious patterns detected.