AgentWard Sanitize

• Outcome: Pass
• Latest audit: May 28, 2026 at 9:37 AM UTC
• Version:
  1.0.0

Overview

This is a local PII redaction skill with disclosed file handling; the main caution is protecting the optional entity-map file that can contain original PII.

Install only if you want a local tool that reads files you choose and creates sanitized copies. Treat any generated *.entity-map.json file as highly sensitive because it can contain the original PII values; store it carefully or delete it when no longer needed.

SkillSpector

By NVIDIA

Vulnerability Patterns

• 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
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 88% confidence
• Finding: The skill declares no permissions, but its documented and referenced functionality clearly requires reading input files and writing sanitized output files. This creates a mismatch between declared capabilities and actual behavior, which can undermine agent sandboxing, policy enforcement, and user trust because file access may occur without explicit permission disclosure.

VirusTotal

65/65 vendors flagged this skill as clean.