Security audit

Parallax

Security checks for vulnerabilities and agentic risk

Overview

This is a reasoning-only decision review skill with no code execution, network access, credential use, or persistence.

Install this if you want the agent to apply a structured five-perspective review to significant product or engineering decisions. Expect it may be invoked in more design discussions than a narrowly scoped architecture-only skill, but the artifact is reasoning-only and does not perform actions on your system.

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 (1)

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The 'When to Invoke' section uses triggers like 'During design reviews,' 'When stakeholders disagree,' and when a decision 'feels right,' which are common and loosely bounded situations rather than specific activation phrases or constraints. Because the file does not provide exclusions or clearer scope limits, the skill could be invoked more broadly than intended in everyday engineering conversations.

Static analysis

No suspicious patterns detected.