ClawHub
Back to skill

Security audit

Using Agent Skills Router

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small routing guide that helps an agent choose an appropriate skill or workflow and does not introduce executable code, persistence, credentials, or hidden data access.

Installers should treat this as a convenience router for choosing skills and workflows. It does not appear to add code execution or hidden access, but users should still rely on the underlying selected skills and normal confirmation practices for risky edits, destructive commands, credentials, production systems, or releases.

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
  • 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 (1)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 48)May include surrounding context.

md
- Do not turn small fixes into a full spec-plan-build-review cycle.
- Do not load long reference files just to decide whether a skill applies.
- Do not role-play multiple agents when the runtime cannot actually isolate their work.
- Do not ask the user which skill to use when local context makes the choice clear.

Static analysis

No suspicious patterns detected.