ClawHub
Back to skill

Security audit

Close Loop

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed end-of-session workflow that can make local repo and memory changes, but its behavior is coherent with its stated purpose and external actions are gated.

Install only if you want an agent to actively wrap up sessions by making local commits, organizing files, and writing scoped memory. Use dry-run mode first for sensitive repositories, and confirm project policy before allowing pushes, deploys, or publish actions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
84% confidence
Finding
The README advertises trigger phrases like 'wrap up', 'close session', and 'end session', which are common conversational phrases and can be invoked unintentionally during ordinary interaction. Because this skill performs end-of-session actions such as shipping state, memory consolidation, and publish/deploy gating, accidental activation could initiate sensitive workflow steps at the wrong time.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly directs moving user document files to a different location without any stated requirement for user confirmation, preview, or rollback. In an end-of-session automation context, this can unexpectedly modify a user's filesystem, break references/workflows, or relocate important documents based on imperfect heuristics.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill instructs committing changes and potentially pushing to remotes, but it does not require a clear user warning, confirmation step, or review of what will be sent. This is dangerous because it can permanently alter repository history and exfiltrate sensitive or unintended changes to remote services if policy checks are incomplete or misapplied.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.