ClawHub
Back to skill

Security audit

Unreasonable Hospitality

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Markdown-only product design guide with broad but disclosed hospitality-oriented UX guidance and no hidden execution, credential use, or persistence.

Install this if you want the agent to apply a warm, hospitality-focused lens to UX, copy, onboarding, feedback, and product review. Because the trigger description is broad, users who prefer tightly scoped skills should invoke it explicitly only for product-experience work.

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
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill description contains very broad trigger phrases such as product design, reviewing quality, writing copy, building assistants, onboarding, feedback loops, and making things feel better or more human. In an agentic system, such expansive invocation cues can cause the skill to activate across a wide range of ordinary tasks, increasing the chance it influences outputs outside its intended niche and crowds out more appropriate, specialized guidance.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% 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 · references/principles.md (reported line 34)May include surrounding context.

md
### 5. Read and React (Improvisational Hospitality)
- Listen → observe → interpret → create → deliver
- Don't wait for feedback forms. Read real-time signals.
- Build systems that allow team members to act on observations without asking permission
- At EMP: servers overheard conversations and created spontaneous personalized moments

### 6. Earned Informality

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% 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 · references/principles.md (reported line 92)May include surrounding context.

md
### 5. Read and React (Improvisational Hospitality)
- Listen → observe → interpret → create → deliver
- Don't wait for feedback forms. Read real-time signals.
- Build systems that allow team members to act on observations without asking permission
- At EMP: servers overheard conversations and created spontaneous personalized moments

### 6. Earned Informality