ClawHub
Back to skill

Security audit

Workspace Trash

Security checks across malware telemetry and agentic risk

Overview

The skill appears to be a deletion/trash helper, and the only surfaced concern is broad activation wording rather than hidden or unrelated behavior.

Install only if you want the agent to assist with file deletion or trash operations. For safety, require explicit confirmation before any delete or trash-empty action, and verify the exact paths before approving.

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 (1)

Vague Triggers

Medium
Confidence
83% confidence
Finding
The trigger language is very broad: it activates on any file deletion request and on generic rm/delete/remove wording. Overly broad matching can cause the skill to intercept unrelated requests or override user intent, increasing the chance of unintended destructive or workflow-altering behavior, especially when combined with operations like trash:empty.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.dangerous_exec

Shell command execution detected (child_process).

Critical
Code
suspicious.dangerous_exec
Location
scripts/trash.sh:208