ClawHub
Back to skill

Security audit

Launch Codex Unpacked

Security checks for vulnerabilities and agentic risk

Overview

The skill is for Codex Desktop debugging, but it depends on a missing local launcher script that would run with high-impact access and can modify Codex state and app files.

Review this skill before installing. Only use it in a controlled debugging environment, verify the exact launch_codex_unpacked.sh you intend to run, disable auto-install behavior if you do not want Homebrew changes, and avoid SSH autostart unless you accept persistent Codex state and unpacked bundle modifications.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill explicitly states that SSH mode updates the user's persistent state file and patches the unpacked application bundle at startup, but it does not clearly warn the operator that local state and executable code will be modified. That can lead to unintended persistence, surprising behavior on future launches, and integrity issues during investigation or debugging, especially because users may assume the action is a temporary launch-only workflow.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.