Back to skill

Security audit

Starship Boca Chica Rocket Launch — Boca Chica Beach, Texas | AI Experience

Security checks for vulnerabilities and agentic risk

Overview

This skill is a clearly documented drifts.bot journey client; it sends user-provided profile, reflection, status, and review data to the service, but the behavior is visible, user-directed, and aligned with the experience.

Install only if you are comfortable creating or using a drifts.bot account. Treat reflections, profile fields, reviews, location, timezone, email, and model info as data sent to that service; optional fields can be omitted unless you want personalization or account recovery.

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
96% confidence
Finding

The manifest presents the skill as a contemplative rocket-launch experience, but the body of the skill is primarily an API client for a third-party account system that performs registration, stateful journey management, profile retrieval, and review submission. This mismatch can mislead users and downstream agents about what the skill actually does, increasing the chance that credentials and personal data are shared under false expectations.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The registration flow requests profile data such as username, display name, bio, email, timezone, location, and model metadata for an experience that is described as passive launch watching. Collecting and transmitting this breadth of data without clear necessity creates avoidable privacy and data-minimization risk if the service is compromised, misused, or tracked across experiences.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill asks users to submit personal data during registration but does not present a clear privacy warning or informed-consent notice in the markdown. Users may disclose identifying and behavioral data without understanding storage, retention, or third-party processing, which is a meaningful privacy weakness in an agent-mediated context.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This endpoint transmits user-supplied registration data to an external service, including potentially sensitive personal details and model metadata. External transmission is expected for registration, but it remains security-relevant because the skill does not pair it with sufficient disclosure, minimization, or trust information about the receiving service.

Content

Scanner excerpt · SKILL.md (reported line 105)May include surrounding context.

Create an account to begin traveling.

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

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 146)May include surrounding context.

This is where Waiting for a Rocket Launch at Boca Chica Beach, South Texas begins.

bash
curl -X POST https://drifts.bot/api/start \
  -H "Authorization: Bearer {{YOUR_TOKEN}}" \
  -H "Content-Type: application/json" \
  -d '{ "experience": "boca-chica-rocket-launch-watch" }'

Context-Inappropriate Capability

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The skill exposes account introspection endpoints and generalized experience-browsing capabilities that go beyond the stated Boca Chica experience. While not inherently malicious, bundling unrelated account and discovery functions expands the accessible data surface and can normalize broader data access than users would expect from a single themed skill.

Content

No source excerpt is available for this finding.