ClawHub

MBTI

Security checks across malware telemetry and agentic risk

Overview

This skill is a personality diagnostic, but it asks for persistent agent-configuration changes without enough user control or rollback guidance.

Install only if you intend to supervise it closely. Treat it as read-only unless the agent shows the exact SOUL.md diff, explains the scope, creates a backup, and gets explicit approval before writing any persistent configuration changes.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger phrases in the skill description are broad enough to match routine requests like testing or optimizing an agent, which can cause the skill to activate unexpectedly. Because this skill's later stages include diagnosis and configuration changes, over-broad triggering increases the chance of unintentional execution of higher-risk behavior.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill explicitly states that Stage 5 will automatically modify SOUL.md, but the document does not present a clear warning, consent gate, or limitation on when file changes occur. Automatic configuration editing can alter agent behavior persistently, creating integrity and safety risks if invoked unintentionally or based on flawed analysis.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The usage example instructs users to run a shell script for the full diagnosis flow without warning that it may perform automated configuration changes. This creates a significant risk that a user will execute the script expecting analysis only, while it may silently reach the modification stage and persistently change agent configuration.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal