Seedream AI Image Generator

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:23 PM UTC
• Version:
  1.0.2

Overview

This is a straightforward Media.io Seedream image-generation skill, but users should know it sends prompts, image URLs, and an API key to Media.io and may use account credits.

Install only if you trust Media.io with the prompts, reference image URLs, task IDs, and account credit lookups you submit. Use a revocable API key, monitor credit usage, avoid private/internal image URLs or sensitive prompts, and install `requests` from a trusted Python package source.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 87% confidence
• Finding: The skill metadata does not explicitly declare permissions, yet the documented usage clearly relies on reading environment variables, local files, and making outbound network requests. This is dangerous because reviewers and users may underestimate the skill's operational reach, especially where secrets and external data transfer are involved.

Tp4

High

• Category: MCP Tool Poisoning
• Confidence: 91% confidence
• Finding: The skill is presented as a narrowly scoped Seedream image generator, but the referenced router and API definition file appear to support broader Media.io API invocation, including non-image endpoints. This scope mismatch is risky because a user or orchestrator may grant trust, secrets, or execution based on the narrower description while the implementation can perform materially different actions.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill encourages sending prompts and image URLs to Media.io but does not prominently warn that this data is transmitted to an external third-party service. This can expose sensitive user content, proprietary prompts, or internal URLs without clear consent, increasing privacy and data-handling risk.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The image-generation documentation explicitly instructs sending user-supplied image URLs and an API key to an external service without any privacy or data-handling notice. This is dangerous because users may provide private image URLs or sensitive content without understanding that it will be transmitted to Media.io and potentially fetched by that third party.

Missing User Warnings

Low

• Confidence: 81% confidence
• Finding: The credits and task-result documentation describes authenticated requests that send account-linked information to an external API, but it omits any warning that user/account metadata leaves the local environment. Although the data involved is limited, the lack of disclosure creates a transparency and privacy risk for users and operators.

VirusTotal

66/66 vendors flagged this skill as clean.