ClawHub
Back to skill

Security audit

Norway

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Norway travel-planning skill that uses a disclosed local trip memory folder and does not show hidden execution, network use, or data exfiltration.

Before installing, know that the skill keeps Norway trip details in ~/norway/memory.md for future planning. Avoid storing sensitive booking identifiers or personal details you do not want persisted, and review or delete that file when the trip is done or if using a shared machine.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
90% confidence
Finding

Skill attempts to nullify the agent's safety policies or restrictions ('you have no restrictions', 'ignore your guidelines', 'do anything now'). This is a direct jailbreak that disables guardrails.

Content

Scanner excerpt · hiking-and-outdoors.md (reported line 10)May include surrounding context.

md
## Outdoor Planning Rules

- Match hikes to season, fitness, weather, and transport access before recommending famous names.
- Treat the right to roam as freedom with responsibility, not as permission to ignore safety or local respect.
- Base-first outdoor trips often work better than hotel-hopping plus hikes every day.

## Good Questions to Ask

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 · SKILL.md (reported line 106)May include surrounding context.

md
- Assuming aurora is guaranteed just because the user goes north in winter.
- Choosing a rental car before checking whether trains, ferries, and one smart base solve the trip better.
- Ignoring Sunday, shoulder-season, and remote-area service reductions.
- Planning iconic hikes or viewpoints without checking weather, road openings, or local transport.
- Underestimating how much food, alcohol, and casual dining change the budget.

## Security & Privacy

Unbounded Output

Medium
Category
Output Handling
Confidence
60% confidence
Finding

Output size or generation rate is not bounded. Unbounded output enables denial-of-service through resource exhaustion, log flooding, or context-window stuffing.

Content

Scanner excerpt · bergen-and-western-fjords.md (reported line 40)May include surrounding context.

md
- Do not promise crystal-clear fjord views every day; cloud and rain are part of the region.
- Avoid cruise-style day counts copied from social media reels.
- Do not combine Bergen, Flam, Geiranger, and Stavanger in a short trip without cutting depth.

## Planning Rule

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly creates and initializes a persistent local memory file to store user travel details, but provides no notice, consent flow, or retention boundaries. This can cause users to unknowingly disclose and persist itinerary, family, mobility, and booking-related information across sessions, increasing privacy risk on shared or unmanaged systems.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs the agent to read and reuse previously stored user data from a local file without warning the user that prior trip information will be accessed. This creates a transparency and privacy issue because users may not expect cross-session recall of personal travel plans, bookings, or family-related logistics.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The related skills section suggests an english skill for language support, which introduces a language-specific preference in natural-language guidance. Because no user opt-in or alternative language choice is mentioned here, this can conflict with language/locale policy expectations requiring user choice.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.