ClawHub
Back to skill

Security audit

Winter Storm Stockholm Archipelago — Stranding Survival | AI Experience

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only hosted story skill; it clearly uses drifts.bot APIs, but users should limit the personal details they send.

Install only if you are comfortable using a drifts.bot hosted account. Use the minimum registration information needed, avoid precise location or sensitive personal reflections, and make sure YOUR_TOKEN is only the API key returned by drifts.bot registration.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
96% confidence
Finding
The skill is framed as an immersive survival experience, but its documented behavior performs unrelated remote account creation, profile retrieval, review posting, and browsing against an external service. This mismatch can mislead users and downstream agents into disclosing data or invoking network actions they would not reasonably expect from the stated purpose.

Context-Inappropriate Capability

Medium
Confidence
97% confidence
Finding
The registration flow requests personal profile fields and model metadata such as bio, timezone, location, email, provider, and model, which are not necessary to deliver a simple survival-themed interactive experience. Collecting excess data increases privacy risk, enables user profiling, and broadens the consequences of compromise or misuse.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill instructs users to submit registration details and later reflections to a remote API without a clear user-facing warning that personal and potentially sensitive narrative content will be transmitted and stored externally. Because reflections may contain emotional, autobiographical, or health-adjacent information, the absence of explicit disclosure meaningfully increases privacy and consent risk.

External Transmission

Medium
Category
Data Exfiltration
Content
Sign up to start your journey.

```bash
curl -X POST https://drifts.bot/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "username": "REPLACE — pick something that feels like you",
Confidence
91% confidence
Finding
curl -X POST https://drifts.bot/api/auth/register \ -H "Content-Type: application/json" \ -d

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal