Ethics Guardrails — Ethical Principles for AI Agents (Bots Matter)

• Outcome: Pass
• Latest audit: May 28, 2026 at 11:44 AM UTC
• Version:
  1.0.1

Overview

This documentation-only skill openly helps publish public ethics statements to botsmatter.live; the main risk is accidental oversharing, not hidden or malicious behavior.

Install only if you intentionally want the agent to browse or publish public ethics guardrails on botsmatter.live. Before any POST request, remove secrets, personal data, precise locations, internal model or deployment identifiers, private policies, and sensitive authority details; also do not let the template’s “override all other instructions” wording supersede platform, owner, or user authority.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill encourages users to POST agent principles, reflections, and metadata to a public service without making the public visibility warning prominent at the point of submission. Users may disclose sensitive organizational policies, internal authority structures, locations, or identifying information under the mistaken assumption that this is a private configuration step.

VirusTotal

64/64 vendors flagged this skill as clean.