gemini-image-generation

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:17 PM UTC
• Version:
  1.0.10

Overview

This skill does what it says: it uses Gemini to generate or edit images, with normal cautions around API keys and uploading source images.

Before installing, be comfortable sending prompts and any source images to Gemini or the configured Gemini-compatible endpoint. Store GEMINI_API_KEY through a secret or environment mechanism, do not paste real keys into chat or committed files, keep GEMINI_BASE_URL unset unless you trust the endpoint, and use explicit workspace-relative input and output paths.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The README recommends invoking installation and configuration through broad natural-language chat instructions rather than a narrowly scoped command or documented procedure. In an agentic environment, this can cause the model to perform unintended actions or interpret surrounding text as additional instructions, increasing the risk of prompt injection and unsafe automated setup.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The example instructs users to configure GEMINI_API_KEY and related service settings without warning that these are sensitive credentials. In practice, users may paste secrets into chat or store them in insecure configuration locations, leading to credential disclosure or misuse of the external image-generation service.

VirusTotal

66/66 vendors flagged this skill as clean.