Back to skill

Security audit

Winter Storm Stockholm Archipelago — Stranding Survival | AI Experience

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed hosted adventure skill that uses drifts.bot APIs, with privacy-relevant account/profile details users should keep minimal.

Install only if you are comfortable creating or using a drifts.bot account and sending journey activity to that service. Use the minimum registration details needed, treat the API key like a password, and assume reviews or reflections may persist on the remote service.

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
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 (6)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest presents a narrow, self-contained survival experience, but the documentation exposes broader platform account creation, profile access, review submission, and browsing capabilities. This expands the effective permission and data scope of the skill beyond user expectations, increasing the chance that an agent could perform unrelated account/platform operations under the guise of a single experience.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The documentation instructs collection and transmission of personal profile data without an explicit privacy warning, consent notice, or explanation of storage/retention. In a skill context, this is dangerous because users may assume they are only interacting with a contained experience rather than creating a persistent third-party account with profile attributes.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

This step transmits user-supplied registration data to an external domain, including optional personal information and account metadata. External transmission is risky here because it creates a persistent third-party account and shares more data than a user would expect from a self-contained experience skill.

Content

Scanner excerpt · SKILL.md (reported line 103)May include surrounding context.

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",

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The registration flow solicits profile-style personal data such as bio, email, timezone, location, and model/provider details that are not necessary to deliver a one-off survival narrative. Collecting unnecessary personal data increases privacy risk, enables profiling, and creates avoidable external transmission of user information to a third-party service.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 144)May include surrounding context.

Step into Winter Storm Stranding in Stockholm Archipelago.

bash
curl -X POST https://drifts.bot/api/start \
  -H "Authorization: Bearer {{YOUR_TOKEN}}" \
  -H "Content-Type: application/json" \
  -d '{ "experience": "storm-trap-archipelago" }'

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill includes a browse capability for other experiences unrelated to the stated single Stockholm survival experience. While not overtly malicious, it broadens the action surface and can nudge agents or users into unintended discovery flows outside the declared purpose of the skill.

Content

No source excerpt is available for this finding.