Back to skill

Security audit

MiniMax GIF Sticker Maker

Security checks for vulnerabilities and agentic risk

Overview

This sticker-making skill has a coherent creative purpose, but it would process user photos and facial likenesses through external MiniMax services using helper files that are not included for review.

Install only if you are comfortable with photos, including faces, being processed by external MiniMax services. Before use, verify the missing referenced scripts and dependency files from the publisher, confirm exactly what data they send, and get explicit consent from anyone whose likeness appears in uploaded images.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger list includes very common terms like 'GIF', 'cartoon', 'emoji', and 'sticker', which can cause the skill to activate in unrelated conversations. Over-broad activation can route ordinary user requests into a workflow that uploads user photos to external APIs, creating privacy and consent risks beyond what the user intended.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill instructs sending user photos to MiniMax image/video generation APIs but does not prominently warn the user beforehand that their images will leave the local environment. Because the workflow may process faces and preserve facial likeness, the missing disclosure undermines informed consent and increases privacy, biometric, and third-party data handling risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

Forcing automatic language detection and prohibiting mixed-language output removes user control over captions and may mis-handle multilingual users. In this skill the primary harm is reduced user autonomy and possible incorrect or culturally inappropriate output, rather than direct system compromise.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

Mandating that all outputs follow detected user language instead of explicit user preference can produce unintended results and override clear user needs. In a creative media workflow this is mostly a UX and consent issue, but it can still lead to wrong captions being rendered into generated assets that are costly to regenerate.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
83% confidence
Finding

The instruction requires that all image prompts be in English regardless of the user's language, which is an explicit language constraint. While this may be technically motivated, the file does not explain the necessity or present it as a documented, justified limitation to the user.

Content

No source excerpt is available for this finding.