ClawHub
Back to skill

Security audit

Codex Pet Generator

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Codex pet generator that uses expected local image-processing, optional narrow brand research, and pet package installation steps.

Install this if you want a feature-rich pet generation workflow and are comfortable with it creating run folders, invoking image generation, optionally researching brand cues on the web, and writing the final pet into your Codex pets directory. Review outputs for trademark-sensitive brand requests because the skill intentionally avoids copying logos or readable marks but uses public brand cues.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding
The skill clearly instructs the agent to read and write many files, copy artifacts into workspace and home-directory paths, and package outputs under `${CODEX_HOME:-$HOME/.codex}/pets`, yet it declares no permissions. That mismatch is dangerous because it can bypass user and platform expectations about what the skill is allowed to access or modify, especially given the large number of filesystem operations and cleanup steps.

Context-Inappropriate Capability

Medium
Confidence
82% confidence
Finding
The brand-discovery section adds a web-search and research workflow that materially expands the skill beyond straightforward pet generation into external information gathering. This increases risk because a user invoking a pet-creation skill may not expect outbound research, retrieval of third-party content, or use of external sources that can introduce privacy, compliance, and prompt-injection exposure from searched pages.

Vague Triggers

Medium
Confidence
76% confidence
Finding
The manifest description is intentionally broad, activating on nearly any keyword, brand, object, style, or reference image related to pets, creation, repair, upgrade, validation, packaging, or installation. Over-broad activation is risky because it can cause the skill to run in contexts the user did not clearly intend, which is especially concerning here since the skill performs substantial file operations, packaging, and optional web research.

Static analysis

No suspicious patterns detected.