ClawHub
Back to skill

Security audit

Engineering Mode

Security checks across malware telemetry and agentic risk

Overview

This coding workflow skill is mostly transparent, but it tells agents to run destructive Git rollback commands without a clear user confirmation gate.

Install only if you are comfortable with a skill that may direct an agent to create branches, make checkpoint commits, and perform hard Git rollbacks. Before using it, require the agent to show `git status`, identify the exact branch/checkpoint, summarize what would be lost, and ask for confirmation before any `git reset --hard` or branch deletion.

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
94% confidence
Finding
The skill documents destructive rollback commands such as `git reset --hard` and branch deletion in a generalized recovery path, but does not require explicit user confirmation, preflight status checks, or warnings about uncommitted work loss. Because this skill is meant to shape agent behavior during code-editing tasks, an agent following it could irreversibly delete user changes or branches during routine recovery.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.