ClawHub
Back to skill

Security audit

Milan

Security checks for vulnerabilities and agentic risk

Overview

This is a Milan guidance skill made of static Markdown advice, with no executable code, credential access, background behavior, or hidden data handling.

Install this if you want Milan-specific planning references. Treat prices, transport fares, visa rules, legal requirements, and business setup guidance as starting points and verify current official sources before spending money or making legal, housing, immigration, or business decisions.

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 (6)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The invocation guidance says the skill should be used when the user asks about Milan 'for any purpose,' followed by a very broad list of contexts. This lacks clear boundaries or exclusion conditions, so ordinary mentions of Milan could unintentionally activate the skill.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- Underestimating humidity in summer.
- Planning full outdoor itineraries without rain fallback.
- Choosing housing without checking winter heating quality.

## Practical Packing Rules
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.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- Underestimating humidity in summer.
- Planning full outdoor itineraries without rain fallback.
- Choosing housing without checking winter heating quality.

## Practical Packing Rules
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.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- Underestimating humidity in summer.
- Planning full outdoor itineraries without rain fallback.
- Choosing housing without checking winter heating quality.

## Practical Packing Rules
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.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- Underestimating humidity in summer.
- Planning full outdoor itineraries without rain fallback.
- Choosing housing without checking winter heating quality.

## Practical Packing Rules
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.

Natural-Language Policy Violations

Low
Confidence
82% confidence
Finding
The statement "Basic Italian effort helps significantly in daily life" communicates a language expectation tied to use of the skill content, but the file does not offer any language or locale choice or clarify that this is optional guidance. Under the stated policy, natural-language content that effectively forces or presumes a specific language without opt-in can be a locale-policy concern.

Static analysis

No suspicious patterns detected.