ClawHub
Back to skill

Security audit

厨房战争：红警风RTS

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to deliver a browser RTS game and its test scripts without hidden data access or unsafe installation behavior.

Install if you want a single-file Kitchen War RTS game skill. Before using it in an existing project, check where the agent plans to copy index.html and avoid overwriting a file you care about. Expect local browser storage for game progress and preferences.

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 (3)

Vague Triggers

Medium
Confidence
73% confidence
Finding
The trigger phrases are broad enough to match generic requests for RTS or browser games, which can cause this specific skill to activate in situations the user did not intend. That raises the chance of unintended file-copying, game-specific modifications, or execution of the skill's heavy verification workflow in an unrelated project context.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The skill blends two scopes: maintaining a specific existing Kitchen War artifact and serving generic Red Alert-style game generation requests. That ambiguity can lead the agent to apply Kitchen War-specific copying, editing, and validation behavior when the user may have intended a new or different project, increasing the risk of unintended workspace changes and overbroad execution.

Missing User Warnings

Low
Confidence
84% confidence
Finding
The instructions direct the agent to copy `assets/index.html` into the user's workspace automatically, but do not require user confirmation before modifying local files. Even though the copied content appears benign, silent workspace writes are risky because they can overwrite files, create confusion about provenance, or violate user expectations around consent.

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
scripts/verify_game.js:140