ClawHub
Back to skill

Security audit

Haccp Plan Drafter

Security checks across malware telemetry and agentic risk

Overview

This document-only food-safety drafting skill is aligned with its stated purpose and does not run code or access data by itself.

Before installing, understand that using this skill may involve sharing sensitive facility, recipe, supplier, and process information with the agent. Treat all generated plans as drafts only; a qualified PCQI or HACCP team should validate, review, sign, and approve any plan before operational, regulatory, or audit use.

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)

Harmful Content Injection

Critical
Category
Prompt Injection
Content
### 4. Hazard analysis (Principle 1)

Use the Codex / NACMCF style. For each process step, work the user through:

1. Biological hazards reasonably foreseeable (e.g., Salmonella, Listeria monocytogenes, E. coli O157:H7 / STEC, C. botulinum, Cronobacter, norovirus, parasites).
2. Chemical hazards reasonably foreseeable (mycotoxins, heavy metals, pesticide residues, undeclared allergens, food-additive overages, processing-aid residues, radionuclides, sanitizer residues).
3. Physical hazards reasonably foreseeable (metal, glass, hard plastic, wood, stones, bone).
4. **Radiological** hazards if applicable.

For each candidate hazard, capture:
Confidence
70% confidence
Finding
This content may contain harmful instructions that could cause physical harm if followed. CRITICAL: Review carefully before use.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.