ClawHub

Research Tracker

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent local research-agent tracking helper, but users should trust the external CLI source and avoid storing sensitive research content in its persistent database.

Install only if you trust the 1645labs Homebrew tap or Go repository, and prefer reviewing or pinning a release before important use. Keep research tasks narrowly scoped, monitor heartbeats and stop signals, and do not log secrets or highly sensitive material unless you are comfortable retaining it in the local SQLite database.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger description is overly broad and includes generic phrases like research projects, coordination, investigation, and oversight, which could cause the skill to activate for many unrelated requests. In an agentic system, overbroad routing can inappropriately introduce persistent state management and autonomous sub-agent workflows into contexts where the user did not request them, increasing the chance of unintended data retention or agent spawning.

Missing User Warnings

Low
Confidence
88% confidence
Finding
The skill documents persistent SQLite storage of project state, instructions, and agent activity but does not prominently warn users that operational data will be retained on disk. Because the tool tracks autonomous agent behavior and prompts/context snippets, users may unknowingly store sensitive research content, instructions, or metadata locally, creating privacy and forensic exposure.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal