ClawHub

Auto Improver Pro

Security checks across malware telemetry and agentic risk

Overview

This appears to be a disclosed local skill-management helper that can propose persistent skill changes, with approval gates before writing.

Install only if you want an agent to periodically evaluate whether conversations should become reusable local skills. Review proposed diffs carefully before approving, because approved changes can persist and shape future agent behavior.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly promotes autonomous optimization, feedback collection, pattern extraction, and automatic execution of changes, but it does not clearly warn users that it may modify skills or process execution/feedback data. In a skill that exposes Write, Bash, and Exec capabilities, this creates a meaningful risk of unauthorized changes, privacy issues, and unsafe self-modification without informed consent or approval boundaries.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrases are overly broad and can activate the skill in loosely related contexts such as generic version upgrades, performance reviews, or weakness identification. For an autonomous self-improving skill, ambiguous activation boundaries increase the chance of unintended execution, unnecessary data processing, and recursive or excessive optimization behavior in situations where the user did not explicitly request it.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal