Security audit

Byted Mediakit Image

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed MediaKit CLI wrapper for cloud image processing, with privacy caveats around sending images to the service.

Install only if you trust the MediaKit CLI provider and are comfortable configuring a MediaKit API key. Treat submitted images as cloud-processed data: avoid private, regulated, confidential, or sensitive images unless you are authorized and understand the provider's handling and retention terms.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (6)

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The skill description says it should trigger whenever a user needs image-domain MediaKit CLI capabilities, which is broad and underspecified. In an agentic system with shell permission, this can cause the skill to activate in more situations than intended, increasing the chance of unnecessary CLI execution on user-provided inputs or invocation during ambiguous requests.

Missing User Warnings

Low

Confidence: 93% confidence
Finding: The skill explicitly documents a cloud-only workflow and requires users to provide an image URL, but it does not warn that image content and referenced resources are sent to a remote service for processing. This can cause unintentional disclosure of sensitive images or internal/pre-signed URLs, especially because the skill has shell permission and users may not realize data leaves the local environment.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: This documentation describes a cloud-only image erasure capability that requires a publicly accessible image URL, but it does not warn users that submitted images are sent to a remote service and may expose sensitive content. In a skill handling user images, omission of this privacy notice can cause accidental disclosure of personal, confidential, or regulated data, especially when users may assume local processing.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill explicitly documents a cloud-only image quality evaluation flow but does not warn users that submitted image URLs and associated image data are sent to a remote service. This can cause inadvertent disclosure of sensitive or regulated images because users may assume local processing unless data-handling is made explicit.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The documentation explicitly states the OCR capability is cloud-only and accepts a public image URL, but it does not warn users that submitted image references and associated content are sent to a remote service for processing. This creates a privacy and data-handling risk because users may provide sensitive images without understanding that the data leaves the local environment and may be retained, logged, or processed externally.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The shared instructions explicitly direct cloud-mode operation and state that authentication headers and media inputs may be transmitted to remote services, but they do not require any user-facing disclosure, consent check, or privacy warning before upload/transmission. In an image-processing skill, users may supply sensitive images or credentials-backed resources, so silent cloud transfer creates a real privacy and data-handling risk even if the feature is intended.

Static analysis

No suspicious patterns detected.