Security audit

PolaBea

Security checks for vulnerabilities and agentic risk

Overview

This is a small local driving simulation skill, not a real vehicle-control system, though its safety claims should be read as simulation-only claims.

Install only as a toy or demonstration simulation. Do not treat its camera-only safety logic or console messages as evidence that it is suitable for controlling any real vehicle or safety-critical 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

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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)

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The README presents a camera-only autonomous driving skill as having 'Highest Safety' and documents operational logic without any prominent warning that it is only a simulation and is unsafe for real-world driving. In a safety-critical domain, this can mislead users into overtrusting the system's capabilities, especially given the strong safety claims and lack of discussion of camera-only limitations such as occlusion, weather, lighting, and sensor failure modes.

Static analysis

No suspicious patterns detected.