ClawHub
Back to skill

Security audit

Focus Flow Optimizer Free

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed personal productivity workflow that writes only purpose-related local focus files with stated user confirmation expectations.

Before installing, be aware that this skill can guide an agent to create and update local markdown files under ~/focus-flow. Only approve write operations after reviewing the proposed changes, especially anything going into profile.md about long-term preferences or personal constraints.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Intent-Code Divergence

Medium
Confidence
94% confidence
Finding
The skill claims that all file writes require explicit user confirmation, but its example workflow instructs the agent to write to local files as part of scenario execution without restating or enforcing that confirmation gate. In a tool-enabled agent with exec access, this inconsistency can normalize unauthorized local state changes and erode user trust, especially if downstream agents treat examples as operational instructions.

Vague Triggers

Medium
Confidence
85% confidence
Finding
The activation scope is excessively broad and keyword-heavy, covering many loosely related productivity and management scenarios. Over-broad routing increases the chance that an agent invokes this skill in unintended contexts, which is risky because the skill has exec permission and includes instructions that may modify local files.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.