Skill Frontmatter Doctor

• Outcome: Review
• Latest audit: May 29, 2026, 9:26 PM
• Version:
  1.0.0

Overview

The skill appears broader than a simple metadata repair helper and should be reviewed before installation.

Install only if you intend to use the broader audit/scanning behavior, not just metadata repair. Limit it to specific directories, avoid running it over private or sensitive workspaces, and require explicit approval before file writes or shell commands.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• 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: 89% confidence
• Finding: The skill advertises and enables code-capable behaviors such as file read/write and shell execution, but does not declare any corresponding permissions or constraints in the manifest. This creates a trust and review gap: users and orchestrators may invoke the skill believing it is a low-risk formatting helper when it can access files and run commands.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 95% confidence
• Finding: The declared purpose is narrow frontmatter and metadata repair, but the documented/observed behavior is much broader, including arbitrary directory scanning, content summarization, CSV/TSV inspection, risk-pattern scanning, and behavior driven by an external spec. This mismatch is dangerous because it can cause overbroad access to user data and execution of capabilities users did not meaningfully consent to under the stated purpose.

VirusTotal

63/63 vendors flagged this skill as clean.