CORE Prism (CORE 四维战略透镜)

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:29 PM UTC
• Version:
  1.0.2

Overview

The skill appears to disclose a purpose-aligned use of local user profile context, with no evidence here of hidden execution, exfiltration, or persistence.

Install only if you are comfortable with the skill using USER.md to personalize responses. Keep secrets, credentials, and highly sensitive personal details out of USER.md unless you intentionally want skills to use that context.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• 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: 91% confidence
• Finding: The README explicitly states that the skill reads `USER.md` to adapt output, but it does not disclose scope limits, consent expectations, or any safeguards around what personal or sensitive data may be consumed. In an agent environment, implicit file access to user-context documents can lead to privacy violations, over-collection of data, and unintended propagation of sensitive information into prompts or downstream outputs.

VirusTotal

65/65 vendors flagged this skill as clean.