ClawHub
Back to skill

Security audit

Eo Ability Dream

Security checks across malware telemetry and agentic risk

Overview

The skill is transparent about its self-learning purpose, but it can automatically use real work logs and persist learned behavior without clear controls.

Install only if you are comfortable with a self-learning EO workflow. Before enabling EO-plugin mode, confirm exactly which logs it can read, whether analysis is opt-in, how sensitive content is redacted, where Pattern Library changes are stored, and how updates can be reviewed, audited, and rolled back.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly states that, with the EO plugin, it can automatically analyze real work logs and update a pattern library, but it provides no consent, scoping, retention, redaction, or approval controls. That creates a real privacy and integrity risk: sensitive data from logs may be processed unintentionally, and automatically learned patterns may propagate mistakes or tainted data into future behavior.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal