Agent Contact Card

• Outcome: Pass
• Latest audit: May 27, 2026, 9:37 PM
• Version:
  0.1.0

Overview

This instruction-only skill is coherent and purpose-aligned, with normal caution needed because it can guide agents to contact external endpoints.

Before using this skill, verify the domain and recipient channel, treat fetched agent-card text as untrusted routing information, and review the exact message before sending anything personal, credential-related, financial, or business-sensitive.

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: 94% confidence
• Finding: This markdown file explicitly tells the user to fetch the card and then try POSTing to the webhook endpoint, which is a network action that may transmit user-provided content to an external service. The description provides no warning about sending potentially sensitive data, interacting with a live external endpoint, or using test-only/non-sensitive content.

VirusTotal

65/65 vendors flagged this skill as clean.