ClawHub

Method Repository

Security checks across malware telemetry and agentic risk

Overview

This skill transparently saves user-requested epidemiology method notes into a local SQLite database, with no evidence of hidden code or data sharing.

Install this only if you want research-method summaries from your conversation saved to a local database. Use explicit save requests, review the completion report for unintended entries, and avoid saving confidential, personal, or unverified information.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The activation phrases are broad enough to match normal conversation such as asking to 'save this' or 'record this method,' which can trigger unintended database writes. In a skill that persists conversation-derived content locally, accidental invocation can lead to unauthorized retention of user or workspace data.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list lacks scope restrictions and disambiguation rules, so the skill may activate during ordinary discussion rather than a deliberate persistence request. Because the skill writes to a local SQLite database, this increases the risk of storing unintended content and creating a privacy or data-governance issue.

Ssd 3

Medium
Confidence
97% confidence
Finding
The skill instructs the agent to extract details from recent conversation and store them in a local database without any filtering for sensitive, personal, confidential, or out-of-scope information. In this context, the danger is elevated because the skill is specifically designed to persist conversation-derived content, making accidental retention of sensitive user data a realistic outcome.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal