ClawHub
Back to skill

Security audit

Batch Plan Execute

Security checks across malware telemetry and agentic risk

Overview

This instruction-only skill is a disclosed planning and execution helper whose workspace writes and code changes are purpose-aligned and gated by explicit user execution commands.

Install this if you want an agent to create persistent planning files and, only when clearly instructed, implement those plans in your repository. Use it in version-controlled projects, review generated plans before saying execution phrases like "implement now" or "apply the plan," and be careful in untrusted repositories because validation commands can run project code.

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)

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
The instruction mandates that final plan and execution-progress language follow the source document's dominant language unless the user explicitly asks for translation. This imposes a language choice by default rather than offering the user a language/locale option, which matches the policy's prohibited forced-language pattern.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.