Claude Usage

Security checks across malware telemetry and agentic risk

Overview

This skill locally calculates Claude usage from OpenClaw session logs and does not show hidden networking, exfiltration, or unrelated actions.

Install only if you are comfortable with a local tool reading your OpenClaw session logs, since those logs may include private conversation content. Use --save only if you want it to store reset time, plan, and timezone locally.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 89% confidence
Finding: The skill instructs the agent to run a local Python script, read session data from the user's filesystem, and write saved configuration, but it does not declare any permissions. That creates a transparency and policy-enforcement gap: a user or runtime may not realize the skill needs shell, file-read, and file-write access before it executes, increasing the chance of unintended local data access or modification.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal