ClawHub
Back to skill

Security audit

Agent Health Optimizer

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed OpenClaw health-audit toolkit that reads local agent setup, writes local reports, optionally adjusts cron timing when explicitly run with --fix, and queries ClawHub for skill comparison.

Install only if you are comfortable with a tool that scans your OpenClaw workspace and memory files, saves diagnostic reports under memory/, and contacts ClawHub to compare installed skills. Run cron_optimizer.py without --fix first, review the suggested cron changes, and use --fix only if adding stagger to selected cron jobs is acceptable. Add the weekly cron only if you want ongoing background audits.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
89% confidence
Finding
The skill documentation indicates capabilities to read workspace files, write reports, invoke shell commands via the OpenClaw CLI, and access an external network API, but it does not declare any permissions. This creates a transparency and policy-enforcement gap: users or the platform may treat the skill as lower risk than it actually is, and the combination of shell, file write, and network access increases the blast radius if the underlying scripts are misused or compromised.

Missing User Warnings

Low
Confidence
89% confidence
Finding
The script enumerates locally installed skill directories and transmits each skill slug to the remote ClawHub API, which leaks metadata about the user's installed tooling without explicit consent or a clear disclosure. While the data sent is limited to slugs rather than full file contents, installed-skill inventory can still reveal interests, workflows, or sensitive operational capabilities in an agent environment.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.