Back to skill

Security audit

karpathy-coding-rules

Security checks for vulnerabilities and agentic risk

Overview

This is a lightweight coding-guidance skill with no executable payload, network behavior, persistence, or hidden authority.

This skill is safe to install if you want coding-workflow guidance. Expect it to influence how the agent plans, limits edits, and verifies coding tasks; review it mainly for language preference and whether you want these principles applied broadly to coding work.

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
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The manifest description says the skill is '触发于写脚本/自动化/修复 bug 等编码任务', which defines activation in very broad terms rather than specific trigger phrases or narrow invocation conditions. Because it also includes '等编码任务', the scope is open-ended and may cause the skill to activate for a wide range of common coding-related requests without clear boundaries.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill is written entirely in Chinese and presents its operating description and guidance in that language, but it does not indicate that language choice is optional or user-selected. Under the policy, forcing a specific language without opt-in is a natural-language policy violation unless the locale restriction is clearly documented and justified.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 10)May include surrounding context.

md
## 核心理念

> "Models make wrong assumptions on your behalf and just run along with them without checking." — Andrej Karpathy

大模型常犯的错误:过度自信、不质疑假设、代码过度工程化、只管写不管对不对。