Persona Pack Builder

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:20 PM UTC
• Version:
  1.0.0

Overview

This skill is a coherent persona/prompt-pack builder whose local file generation is disclosed and purpose-aligned.

Install only if you want a workflow that helps create persona/prompt pack files. Run the optional Python generator in a new empty output folder, review generated prompts before publishing or selling them, and avoid requests that impersonate real people or market a persona as a real individual.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• 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

Findings (2)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 93% confidence
• Finding: The skill explicitly instructs the agent to read local references/templates and to write multiple output files, and even to invoke a local packaging script, yet it declares no permissions. This creates a capability/permission mismatch: users or the platform may believe the skill is non-file-operating when it is designed to perform file I/O, which weakens reviewability, consent, and sandbox enforcement.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The trigger examples are open-ended productization requests without clear scope limits, exclusions, or safety gating. In a skill designed to build and package persona/prompt products, broad activation language can cause the agent to engage on requests that drift into unsafe persona creation, manipulation, or other policy-sensitive prompt packaging without first checking boundaries.

VirusTotal

66/66 vendors flagged this skill as clean.