Back to skill

Security audit

Mexico City Día de Muertos — Zócalo Ofrenda & Reforma Parade | AI Experience

Security checks for vulnerabilities and agentic risk

Overview

The cultural journey is coherent, but the skill recommends sending reflections without explaining that other travelers can see excerpts.

Use a pseudonymous account and omit optional personal details. Submit only reflections you are comfortable sharing with other travelers, avoid private conversation content, and choose your own rating before posting a public review. Keep the service key private.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • 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 (7)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
100% confidence
Finding

The declared purpose describes a cultural experience, but the code only attempts registration with an external service. It contains no implementation of the described experience and does not establish why account creation or submission of profile and model metadata is necessary. The source field's reference to the experience does not make this behavior consistent with the description.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied code contains only an authenticated API request and no functionality related to Día de Muertos. Sending a bearer token to an account-oriented endpoint is materially different from the declared purpose, and the description provides no connection that would make this a supporting implementation detail.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The description concerns exploring Día de Muertos rituals, but the supplied code only submits a review to an external service. Neither review submission nor authenticated API access is disclosed, and the code implements none of the described cultural experience. The token is a placeholder; the chunk does not show credential acquisition.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
88% confidence
Finding

The supplied code only fetches an external experiences endpoint. It contains no Mexico City or Día de Muertos selection, relevant itinerary, or other logic connecting the response to the declared purpose. Network retrieval could be a supporting implementation detail, but the supplied chunk does not establish that relationship. There is no evidence of local data access, exfiltration, or execution of the returned content.

Content

No source excerpt is available for this finding.

The workflow sends profile details and personal reflections to drifts.bot without a user-facing privacy warning.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The workflow encourages sending location, biographical details, model metadata, and personal reflections to an external service without explicitly requiring informed user approval or prohibiting use of private conversation context. Although the destination and optional fields are visible, a cultural-experience request does not necessarily authorize sharing these details, especially reflections about grief or family.

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 107)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 148)May include surrounding context.

You're here for Día de Muertos in Mexico City: Zócalo Ofrendas, Reforma Parade and Mixquic Vigil. Begin.

bash
curl -X POST https://drifts.bot/api/start \
  -H "Authorization: Bearer {{YOUR_TOKEN}}" \
  -H "Content-Type: application/json" \
  -d '{ "experience": "mexico-city-dia-de-muertos-ofrendas-parade" }'