ClawHub

codex-mcp-dev

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed local Codex coding helper that can change project files, but its behavior matches its stated purpose.

Install only if you want a skill that can delegate real coding work to your local Codex setup. Use a narrow --cwd, prefer read-only mode for analysis, review diffs and test results, and avoid --yolo or danger-full-access unless you explicitly need that level of authority.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill clearly instructs the agent to invoke shell commands and operate on local repositories, yet it declares no explicit permissions or capability boundaries. This creates a transparency and policy-enforcement gap: downstream systems or reviewers may treat the skill as lower risk than it is, while it can still trigger file reads and code-modifying command execution.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The description says to use the skill for broadly defined 'real coding work' and many categories of development tasks, which can cause over-invocation in situations where direct, minimal edits would be safer. Overly broad activation increases the chance that an autonomous code-executing workflow is selected unnecessarily, expanding exposure to unintended repository changes or command execution.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The workflow explicitly recommends `ask --full-auto` for implementation, refactor, and test-writing, which authorizes autonomous code modification, but it does not pair that recommendation with an explicit warning to the user that repository files may be changed. In the context of a coding skill with shell access, this makes unintended or insufficiently reviewed modifications materially more likely.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal