Security audit

Karpathy Coding Guidelines

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only coding-guidelines skill that gives cautious development workflow advice without executing code or requesting sensitive access.

Before installing, understand that this skill may influence many coding-related conversations by encouraging slower, more scoped, test-driven work. That is appropriate for its purpose, but users who only want opt-in use may prefer a narrower trigger.

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: 96% confidence
Finding: The trigger scope is broad enough to activate on many generic coding tasks, code reviews, refactors, and bug fixes rather than a narrowly defined intent. This can cause the skill to be invoked unexpectedly, influencing agent behavior in contexts where the user did not explicitly request these guidelines, which increases prompt-surface area and can create instruction conflicts or unnecessary behavioral steering.

Static analysis

No suspicious patterns detected.