ClawHub
Back to skill

Security audit

Smart Router Omni

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed routing helper that inspects installed skill metadata to recommend a skill or short chain, with no executable code or persistence found.

Install this if you want a broad helper that can decide which skill or skill chain should handle a request. Be aware that it may be invoked implicitly and may inspect visible skill metadata and environment indicators to route tasks, so users who prefer explicit skill selection should review that behavior before installing.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill advertises extremely broad invocation criteria such as choosing skills whenever multiple skills overlap, tasks span phases, or local metadata should be inspected before routing. That makes the router likely to trigger on many ordinary requests and potentially intercept tasks that should go directly to more constrained specialist skills, increasing the chance of unsafe delegation, unexpected metadata inspection, and over-collection of context.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The default prompt instructs automatic environment inspection and routing to the 'best skill or skill chain' without clearly constraining scope, authority, or selection criteria. In combination with implicit invocation, this can cause the agent to invoke more capabilities than the user intended, increasing the chance of unintended data access, unsafe chaining, or execution of a higher-risk skill based on ambient environment signals.

Content

No source excerpt is available for this finding.