Back to skill

Security audit

next ai game

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward guide for uploading a browser game to an external game-hosting API, with no hidden code or persistence.

Before installing, understand that using this skill may upload your game files to a third-party service and produce a public game URL. Only upload files you intend to publish, and confirm the gameFolder and destination endpoint before running the curl commands.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill instructs uploading local files to an external domain and publishing a game URL, but it does not clearly warn that data leaves the local environment and may become externally accessible. This can lead users or downstream agents to exfiltrate local workspace files or unintentionally publish proprietary content.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The document first presents an 'Important Rule' that existing files cannot be overwritten and that the interface will error if a file exists. Later, it says overwriting is possible via a general interface with overwrite=true, which contradicts the earlier absolute statement about overwrite behavior and changes the implied operational intent.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

This is an explicit prohibition on a particular language script in natural-language instructions, without explaining that the limitation is purely technical or offering user choice. SQP-3 covers language or locale policy violations when a skill forces or excludes specific languages without opt-in or clear justification.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The instruction 'use English' is a natural-language locale restriction, and the file does not offer a language choice or explain a necessary region-specific justification. Under SQP-3, forcing a specific language without user opt-in can violate language or locale policy.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.