ClawHub
Back to skill

Security audit

SwitchBoard

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed model-routing guide with no executable code, but users should verify its OpenRouter/privacy choices and mismatched project links before relying on it.

Install only if you intend to route agent prompts through OpenRouter models. Add the API key knowingly, avoid sending secrets or private PII to free or unmoderated models, consider narrowing the activation triggers, and verify the publisher/source links because the listed repository and homepage do not clearly match the skill.

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
  • 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 (3)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger list contains broad phrases such as "expensive," "cheap model," and "which model" that are likely to appear in ordinary conversations. This can cause the skill to activate outside its intended scope, potentially influencing model routing decisions and encouraging use of external providers when the user did not explicitly request that behavior.

Natural-Language Policy Violations

Low
Confidence
84% confidence
Finding
The description states the model is "Optimized for Korean with English and Japanese support," which imposes a locale/language preference in natural-language metadata without indicating user choice or explicit opt-in. Under the policy rule, language constraints should either be optional or clearly justified as region-specific.

Context-Inappropriate Capability

Low
Confidence
94% confidence
Finding
The manifest points to a repository, bug tracker, and homepage that appear unrelated to the declared switchboard skill. Even though this does not directly execute code, mismatched provenance metadata is a supply-chain trust issue because users may review the wrong source, file issues in the wrong place, or rely on misleading ownership signals when deciding whether to install the skill.

Static analysis

No suspicious patterns detected.