ClawHub

SwitchBoard

Security checks across malware telemetry and agentic risk

Overview

SwitchBoard is an instruction-only model-routing skill that is generally coherent and disclosed, with some provenance and activation-scope caveats users should notice.

Install only if you want OpenClaw to route some work through OpenRouter and possibly cheaper or free third-party models. Protect the OpenRouter key, set budget limits if available, avoid routing secrets or private data through free/unmoderated models, and be aware that the package metadata links do not clearly match this skill.

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)

Context-Inappropriate Capability

Low
Confidence
93% confidence
Finding
The manifest metadata points users to a trading repository, issue tracker, and homepage that are unrelated to the stated switchboard/model-routing function. This creates a supply-chain trust problem: reviewers and users may be misled about the skill's provenance, and could follow links to an unrelated external project where they may obtain incorrect code, support, or updates.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger list contains broad everyday phrases such as "expensive," "cheap model," and "which model," which may cause the skill to activate in ordinary conversation outside the user's intent. In a routing skill, unintended activation could inappropriately steer the agent toward lower-cost or third-party models, increasing privacy and reliability risk, especially because the skill explicitly discusses sending work through external providers and free models.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal