ClawHub

Oejts Personality Tuner

Security checks across malware telemetry and agentic risk

Overview

This skill transparently scores a personality questionnaire and can save scoped assistant-preference blocks into USER.md and SOUL.md.

Install this only if you want questionnaire-derived interaction preferences stored in your workspace. Run the dry-run first, review the USER.md and SOUL.md managed blocks, and remove those blocks later if you no longer want that personalization.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill instructs the agent to read and modify workspace files (USER.md and SOUL.md) via the `apply` command, but no explicit permissions are declared. That mismatch can cause the skill to obtain file read/write behavior without transparent user-facing permission boundaries, increasing the risk of unintended or overbroad file modification in a sensitive configuration context.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The apply command writes directly to USER.md and SOUL.md in the provided workspace, changing persistent configuration files that affect future assistant behavior. Although a --dry-run mode exists, there is no mandatory confirmation, backup, or explicit warning before modification, so a user or calling agent can unintentionally overwrite existing content or silently alter behavioral guidance.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal