ClawHub
Back to skill

Security audit

Token Watch

Security checks across malware telemetry and agentic risk

Overview

TokenWatch is a local AI cost-tracking skill whose file writes match its stated purpose, with the main caution that usage history and labels are stored on disk.

This is reasonable to install for local token-cost tracking. Avoid putting secrets, client names, or sensitive project details in task labels or session IDs, and treat `.tokenwatch/` plus exported reports as private usage logs that should not be committed or shared without review.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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

Low
Confidence
84% confidence
Finding
The documentation explicitly promotes local-only storage and report export but does not warn that usage records, task labels, session IDs, and cost history may still contain sensitive operational metadata written to disk. This can expose prompt intent, workflow names, or business activity to other local users, backups, logs, or accidentally committed files, especially because the default `.tokenwatch/` path may be overlooked.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.