ClawHub
Back to skill

Security audit

ActivityClaw Plugin Usage

Security checks across malware telemetry and agentic risk

Overview

This is a transparent guide for a local activity-monitoring plugin, with real privacy and supply-chain considerations but no evidence of hidden or malicious behavior.

Install only if you are comfortable with an activity tracker recording agent file activity, shell commands, web actions, messages, and sub-agent sessions into a local SQLite database and showing them in a localhost dashboard. Verify the third-party npm package and repository first, and stop the service when you do not want activity recorded.

SkillSpector

By NVIDIA
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
Confidence
85% confidence
Finding
The skill explicitly promotes broad monitoring of file operations, command executions, web activity, messages, and sub-agent actions, and later notes that this data is stored locally in a SQLite database, but it does not present any privacy warning, consent requirement, retention guidance, or access-control considerations. Even if storage is local, this creates a meaningful privacy and surveillance risk because sensitive operational history may be collected and exposed through the dashboard or database without the user appreciating the scope of tracking.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.