Codex Token Usage

• Outcome: Pass
• Latest audit: Jun 9, 2026 at 6:37 AM UTC
• Version:
  1.0.0

Overview

This skill appears benign: it reads local Codex session logs to calculate token usage and does not show network transfer, persistence, or hidden behavior.

Install only if you are comfortable with a local script reading your Codex session log files to compute aggregate token usage. Use a specific --codex-home path if you want to limit which Codex logs are analyzed.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 89% confidence
• Finding: The skill instructs the agent to execute a local Python script, read local Codex session logs, access environment-dependent paths, and optionally write machine-readable output for downstream automation, yet it declares no permissions. This mismatch is dangerous because users and enforcement systems may assume the skill is non-privileged when it actually performs shell execution and local file access, increasing the risk of unintended data exposure from sensitive usage logs and reducing policy visibility.

VirusTotal

63/63 vendors flagged this skill as clean.