LYGO Champion: SANCORA — Unified Minds

• Outcome: Pass
• Latest audit: May 28, 2026 at 9:44 AM UTC
• Version:
  1.0.0

Overview

This is a persona-style advisory skill with only disclosed, local package self-checking and no evidence of hidden data access or unsafe automation.

Install this only if you want the SANCORA persona and its opinionated advisory style. The local helper scripts appear limited to package verification, but treat the linked LYGO-MINT verifier as a separate external tool and avoid giving it secrets or private data unless you independently trust it.

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: 88% confidence
• Finding: The skill declares itself as a passive persona helper, but its documented behavior requires reading local package files such as references/canon.json and verifier_usage.md. Undeclared file-read capability expands the skill's effective privilege surface and can mislead operators, making review and sandboxing less reliable.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 93% confidence
• Finding: The skill is presented as an advisory persona, but it also performs integrity checks, parses local metadata, validates a hardcoded external verifier reference, and emits a hash value from canon.json. This hidden operational behavior creates a trust gap: users may invoke what appears to be harmless guidance while actually triggering verification logic and disclosure of internal package metadata, which is especially risky in a persona-branded skill that encourages trust.

VirusTotal

64/64 vendors flagged this skill as clean.