Security audit

open-ugc-kit

Security checks for vulnerabilities and agentic risk

Overview

This skill creates video-ad prompts through a clearly described workflow, with rendering controlled by the user.

Share only photos and product details you are authorized to submit to the named AI services, and review their data-use terms. Review the installer and repository revision before running the unpinned installation command.

A.I.G

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

SkillSpector

Vulnerability Patterns

Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (9)

The generation workflow sends the scene prompt and first-frame image to a third-party service without a privacy warning.

Medium

Category: Output Handling
Confidence: 91% confidence
Finding: The quick-start workflow instructs users to run image-to-video generation on ezugc.ai using their still and scene prompt; this is an operational step, not merely a tool listing. The README does not warn that these inputs are shared with an external service or advise against submitting confidential product details or personal imagery without appropriate authorization.
Content: No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium

Category: MCP Rug Pull
Confidence: 92% confidence
Finding: The documented installation command executes an npm CLI without pinning its version and installs skill content from an unpinned repository reference. Although this is not an MCP server, mutable dependencies create a supply-chain exposure if an upstream release or repository is compromised; exploitation requires someone to run the command.
Content: No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: file_write.

Medium

Category: MCP Least Privilege
Confidence: 70% confidence
Finding: Without declared permissions the skill's intent is opaque and cannot be validated.
Content: No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium

Category: MCP Rug Pull
Confidence: 92% confidence
Finding: The installation command can fetch and execute an unpinned npm CLI and obtain mutable skill content. This is a conditional supply-chain risk rather than an MCP configuration issue: a compromised upstream dependency could affect users who execute the command.
Content: No source excerpt is available for this finding.

The workflow directs users to submit reference and product images to external AI services without a privacy warning.

Medium

Category: Output Handling
Confidence: 93% confidence
Finding: These steps instruct users to submit screenshots and product photos to a text model and use ezugc.ai for generation. The workflow also permits a creator reference photo, but never warns that submitted images may expose personal likenesses, private background details, or confidential product information to third-party services; the likeness-rights guardrail does not address this privacy risk.
Content: No source excerpt is available for this finding.

The generation workflow sends supplied images to third-party services without a privacy warning.

Medium

Category: Output Handling
Confidence: 93% confidence
Finding: The workflow instructs users to process an ad screenshot and product photo, then generate a remake and video on ezugc.ai. Although the service is named, the document does not warn that submitted images and prompts are shared with external services or advise against submitting sensitive or confidential material.
Content: No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium

Category: MCP Rug Pull
Confidence: 92% confidence
Finding: This source documents an installation command that leaves both the executable npm CLI and skill repository revision unpinned. It is not an MCP server reference, but it exposes installations to unexpected upstream changes or supply-chain compromise when the command is executed.
Content: No source excerpt is available for this finding.

The workflow instructs users to submit product and optional creator photos to an external generation service without a privacy warning.

Medium

Category: Output Handling
Confidence: 90% confidence
Finding: The workflow directs users to attach reference photos and generate images and video on ezugc.ai. It does not warn that these assets, potentially including an identifiable person's photo or confidential product imagery, will be shared with a third-party service.
Content: No source excerpt is available for this finding.

The workflow instructs users to submit imagery to external generation services without a privacy disclosure.

Medium

Category: Output Handling
Confidence: 90% confidence
Finding: The workflow accepts user-provided product and packaging descriptions or photos, directs image generation, and then instructs submission of the resulting still to Seedance 2.5. It does not warn that these materials will be processed by external services, potentially exposing confidential product details or identifiable imagery.
Content: No source excerpt is available for this finding.