ClawHub
Back to skill

Security audit

Advocatus

Security checks across malware telemetry and agentic risk

Overview

This skill is a local doctrine-opposition tracker and evaluator with disclosed, narrow file writes and no evidence of hidden data access or exfiltration.

Install this only if you want a local adversarial-review registry for doctrines or rules. Treat edits to the opposition registry and DOCTRINES dictionary as governance changes, and review generated results files if you use them in downstream decisions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
79% confidence
Finding
The skill claims it only records opposition and does not make decisions, yet the documented behavior includes doctrine scoring, clearance state changes, and persistent result generation. This mismatch can mislead operators and downstream agents into granting broader trust than warranted, causing unauthorized evaluative actions or repository state changes under the guise of a passive advisory skill.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.