ClawHub
Back to skill

Security audit

Limited Info Skill Verify

Security checks for vulnerabilities and agentic risk

Overview

This is an instruction-only skill for testing other skills with a single minimally informed subagent, with no executable code or hidden data handling.

Install this if you want a helper for testing skill robustness. Use it only with target skills and artifacts you are comfortable having a subagent exercise, and consider disabling implicit invocation if you want strictly manual control.

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
Confidence
89% confidence
Finding
The skill description and default prompt define invocation in very broad terms, allowing a subagent to be tasked with testing arbitrary other skills with only minimal input and no explicit scope limits. This can cause unintended or risky delegation behavior, especially when combined with implicit invocation, because the agent may activate in contexts the user did not clearly intend and may evaluate untrusted skills or artifacts without sufficient guardrails.

Vague Triggers

Medium
Confidence
95% confidence
Finding
Enabling implicit invocation without narrowing conditions means the platform may automatically select this skill in loosely related contexts, even when the user did not explicitly request a minimal-information subagent test. Because this skill is designed to exercise other skills under sparse instructions, accidental activation increases the chance of unsafe delegation, misuse of external artifacts, or unintended testing of sensitive capabilities.

Static analysis

No suspicious patterns detected.