Fire Protection Narrative Report Drafter

Security checks across malware telemetry and agentic risk

Overview

This is a non-executable drafting skill, but it may generate fire-protection engineering judgments that should only come from licensed professionals.

Install only if a licensed fire-protection professional will review and approve every code-path, hydraulic, pump, storage, AMME, and compliance-related statement. Treat the output as a drafting aid, not as engineering analysis or a permit-ready determination.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Description-Behavior Mismatch

Medium
Confidence
89% confidence
Finding
The skill’s guardrails say it only drafts a narrative and must not perform engineering calculations or certify compliance, yet the workflow directs the model to generate allowable-area analyses, summarize hydraulic demand against supply, and propose whether a fire pump or water storage is needed. In this fire-protection context, those outputs can materially influence code path and life-safety design decisions, so users may rely on unreviewed AI-generated engineering judgments as if they were mere drafting assistance.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal