ClawHub
Back to skill

Security audit

RouterBase Model Routing

Security checks for vulnerabilities and agentic risk

Overview

This skill is a RouterBase model-selection aid with disclosed catalog/pricing API lookups and no executable code, persistence, or hidden high-impact behavior.

Before installing, users should understand that the skill may guide agents to query RouterBase catalog or pricing APIs using an existing ROUTERBASE_API_KEY. Treat model IDs and prices as live data to verify, and consider narrowing invocation text if you only want the skill used for explicit RouterBase requests.

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
89% confidence
Finding

The skill description is very broad and includes many common model-selection and architecture-planning scenarios, which can cause the skill to be invoked in contexts beyond its narrow intended use. Over-broad auto-invocation increases the chance of irrelevant or premature routing guidance being inserted into conversations, which can mislead downstream decisions or expose unnecessary external-call behavior.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The default invocation prompt is generic and broadly scoped, telling the agent to use the skill to 'pick RouterBase models for this workload' without defining clear triggering boundaries, authorization limits, or exclusions. In agent ecosystems, overly broad activation language can cause unnecessary or inappropriate skill invocation, which may lead to incorrect routing decisions, expanded data exposure to external services, or unintended influence over model/provider selection.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.