ClawHub
Back to skill

Security audit

Skill Router

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only skill-routing helper that advises which installed skill to use and does not add code execution, persistence, credential use, or hidden data flows.

This skill appears safe to install for users who want help choosing among installed skills. It may occasionally nudge toward a routing choice, so users should still confirm before installing new third-party skills or invoking sensitive downstream capabilities.

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
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • 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)

Direct Prompt Extraction

High
Category
System Prompt Leakage
Content
If an installed skill can complete the task without clearly increasing failure risk or wasted work, do not trigger discovery just because another skill might be more specialized or more novel.
Follow `references/resolution-order.md` when the ranking is not obvious.

## Output rules

Keep routing advice short, concrete, and low-noise.
Confidence
85% confidence
Finding
Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The trigger description uses broad, everyday routing phrases like asking whether a skill exists or which skill to use. In a skill-heavy environment, this can cause the router to activate too often and insert itself into normal workflows, creating unintended delegation decisions and increasing the chance that more sensitive downstream skills are invoked unnecessarily.

Natural-Language Policy Violations

Medium
Confidence
93% confidence
Finding
The example user input and prescribed "Good answer" are written in Chinese, and the file presents them as the expected response pattern without any opt-in or language-selection note. This can violate language/locale policy because it implicitly constrains the assistant to respond in a specific language rather than matching or offering user preference.

Static analysis

No suspicious patterns detected.